Changes

62 days of records

Each snapshot is compared with the previous day: models added and removed, scores and prices moved.

Jul 23, 2026 — Sep 26, 2026

09-26

Sep 26, 20261 changes

ScoreApodex 1.1ApodexIntelligence index30.4→26.4-13%

09-25

Sep 25, 20269 changes

AddedDeepSeek V4.1 Flash (Non-Reasoning)DeepSeekIntelligence index 24.7
AddedGLM-5.3 (low)Z AIIntelligence index 34.3
RemovedAgnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B)Sapiens AI
RemovedAgnes 2.5 Pro BetaSapiens AI
RemovedAgnes 3.0 FlashSapiens AI
PriceGemma 4 26B A4B (Reasoning)GoogleInput price$0.120→$0.100-17%
PriceMuse Glimmer (high)MetaInput price$0.350→$0.320-8.6%
PriceMuse Glimmer (high)MetaOutput price$1.50→$1.35-10.0%
ScoreApodex 1.1ApodexAgentic indexFirst measured→28.4

09-24

Sep 24, 202636 changes

AddedMercury 2.5InceptionIntelligence index 12.3
PriceDevstral 2MistralInput price$0→Metric withdrawn
PriceDevstral 2MistralOutput price$0→Metric withdrawn
PriceDevstral Small 2MistralInput price$0→Metric withdrawn
PriceDevstral Small 2MistralOutput price$0→Metric withdrawn
PriceLFM2.5-8B-A1BLiquid AIInput price$0→Metric withdrawn
PriceLFM2.5-8B-A1BLiquid AIOutput price$0→Metric withdrawn
PriceLFM2.5-VL-1.6BLiquid AIInput price$0→Metric withdrawn
PriceLFM2.5-VL-1.6BLiquid AIOutput price$0→Metric withdrawn
PriceLing-3.0-flash-FinInclusionAIInput price$0→$0.070—
PriceLing-3.0-flash-FinInclusionAIOutput price$0→$0.220—
PriceLlama 3.1 Nemotron Instruct 70BNVIDIAInput price$1.20→Metric withdrawn
PriceLlama 3.1 Nemotron Instruct 70BNVIDIAOutput price$1.20→Metric withdrawn
PriceLlama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIAInput price$0.400→Metric withdrawn
PriceLlama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIAOutput price$0.400→Metric withdrawn
PriceLlama Nemotron Super 49B v1.5 (Reasoning)NVIDIAInput price$0.400→Metric withdrawn
PriceLlama Nemotron Super 49B v1.5 (Reasoning)NVIDIAOutput price$0.400→Metric withdrawn
PriceMistral 7B InstructMistralInput price$0.250→$0.150-40%
PriceMistral 7B InstructMistralOutput price$0.250→$0.200-20%
PriceMistral MediumMistralInput price$1.50→Metric withdrawn
PriceMistral MediumMistralOutput price$7.50→Metric withdrawn
PriceMistral Medium 3.1MistralInput price$0.400→Metric withdrawn
PriceMistral Medium 3.1MistralOutput price$2.00→Metric withdrawn
PriceMistral Small (Feb '24)MistralInput price$0.150→Metric withdrawn
PriceMistral Small (Feb '24)MistralOutput price$0.600→Metric withdrawn
PriceMistral Small (Sep '24)MistralInput price$0.200→Metric withdrawn
PriceMistral Small (Sep '24)MistralOutput price$0.600→Metric withdrawn
PriceMistral Small 3MistralInput price$0.100→$0.050-50%
PriceMistral Small 3MistralOutput price$0.300→$0.080-73%
PriceMistral Small 3.1MistralInput price$0.100→$0.110+10.0%
PriceMistral Small 3.1MistralOutput price$0.300→$0.170-43%
PriceMistral Small 3.2MistralInput price$0.100→$0.070-30%
PriceMistral Small 3.2MistralOutput price$0.300→$0.200-33%
PriceNVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIAInput price$0.200→Metric withdrawn
PriceNVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIAOutput price$0.600→Metric withdrawn
PriceTrinity Large ThinkingArcee AIOutput price$0.900→$0.800-11%

09-23

Sep 23, 202627 changes

AddedClaude Opus 5.5 (Adaptive Reasoning, High Effort, Default Fallback)AnthropicIntelligence index 53.6
AddedClaude Opus 5.5 (Adaptive Reasoning, Low Effort, Default Fallback)AnthropicIntelligence index 42.3
AddedClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback)AnthropicIntelligence index 57.6
AddedClaude Opus 5.5 (Adaptive Reasoning, Medium Effort, Default Fallback)AnthropicIntelligence index 51.2
AddedClaude Opus 5.5 (Adaptive Reasoning, Xhigh Effort, Default Fallback)AnthropicIntelligence index 56
AddedGPT-6 Luna (high)OpenAIIntelligence index 32.1
AddedGPT-6 Luna (low)OpenAIIntelligence index 20.9
AddedGPT-6 Luna (max)OpenAIIntelligence index 37.3
AddedGPT-6 Luna (medium)OpenAIIntelligence index 29.5
AddedGPT-6 Luna (Non-reasoning)OpenAIIntelligence index 18.3
AddedGPT-6 Luna (xhigh)OpenAIIntelligence index 33.9
AddedGPT-6 Sol (high)OpenAIIntelligence index 42.8
AddedGPT-6 Sol (low)OpenAIIntelligence index 33.9
AddedGPT-6 Sol (max)OpenAIIntelligence index 47.5
AddedGPT-6 Sol (medium)OpenAIIntelligence index 39.8
AddedGPT-6 Sol (Non-reasoning)OpenAIIntelligence index 28.1
AddedGPT-6 Sol (xhigh)OpenAIIntelligence index 44.1
PriceGLM-5.1 (Reasoning)Z AIInput price$1.20→$1.28+6.7%
PriceGLM-5.1 (Reasoning)Z AIOutput price$4.40→$4.07-7.5%
Pricegpt-oss-20b (high)OpenAIInput price$0.060→$0.070+17%
Pricegpt-oss-20b (high)OpenAIOutput price$0.190→$0.180-5.3%
Pricegpt-oss-20b (low)OpenAIOutput price$0.200→$0.210+5.0%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAOutput price$2.40→$2.50+4.2%
PriceNemotron 3.5 LightningNVIDIAInput price$0.060→$0.070+17%
PriceNemotron 3.5 LightningNVIDIAOutput price$0.200→$0.220+10.0%
PriceNova PremierAmazonInput price$2.50→Metric withdrawn
PriceNova PremierAmazonOutput price$12.50→Metric withdrawn

09-22

Sep 22, 202610 changes

AddedGrok 4.7 (high)SpaceXAIIntelligence index 46.3
AddedGrok 4.7 (xhigh)SpaceXAIIntelligence index 46.4
AddedMiMo-V2.6-ProXiaomiIntelligence index 46.3
PriceLing-3.0-flash-VLInclusionAIInput price$0→$0.070—
PriceLing-3.0-flash-VLInclusionAIOutput price$0→$0.220—
PriceNemotron 3 Nano Omni 30B A3B ReasoningNVIDIAInput price$0.090→$0.200+122%
PriceNemotron 3 Nano Omni 30B A3B ReasoningNVIDIAOutput price$0.360→$1.09+203%
PriceNemotron 3 Super 120B A12B (Reasoning)NVIDIAInput price$0.190→$0.300+58%
PriceNemotron 3 Super 120B A12B (Reasoning)NVIDIAOutput price$0.650→$0.900+38%
ScoreQwen3.8 27B (Non-reasoning)AlibabaIntelligence index22.4→20.2-9.8%

09-20

Sep 20, 2026277 changes

AddedStep 5 PreviewStepFunIntelligence index 43.7
ScoreAgnes 2.5 Pro AlphaSapiens AIIntelligence index27.8→26.8-3.6%
ScoreAgnes 2.5 Pro AlphaSapiens AIAgentic index29.4→Metric withdrawn
ScoreClaude 4.5 Haiku (Reasoning)AnthropicIntelligence index17.6→16.9-4.0%
ScoreClaude 4.5 Haiku (Reasoning)AnthropicAgentic index10.3→8-22%
ScoreClaude 4.5 Sonnet (Reasoning)AnthropicIntelligence index21.2→20.7-2.4%
ScoreClaude 4.5 Sonnet (Reasoning)AnthropicAgentic index17.5→15.8-9.7%
ScoreClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)AnthropicIntelligence index49.7→49.6-0.2%
ScoreClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)AnthropicAgentic index51→50.7-0.6%
ScoreClaude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)AnthropicAgentic index53.3→53.1-0.4%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)AnthropicIntelligence index47→46.8-0.4%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)AnthropicAgentic index47.8→47.1-1.5%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)AnthropicAgentic index58→57.9-0.2%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)AnthropicIntelligence index49.1→48.9-0.4%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)AnthropicAgentic index50.6→50.2-0.8%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)AnthropicAgentic index57.1→57.2+0.2%
ScoreClaude Opus 4.7 (Adaptive Reasoning, Max Effort)AnthropicAgentic index39.5→38.6-2.3%
ScoreClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index42→41.8-0.5%
ScoreClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicAgentic index42.6→41.9-1.6%
ScoreClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicIntelligence index48.2→48.1-0.2%
ScoreClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicAgentic index52.7→52.3-0.8%
ScoreClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicIntelligence index39.8→39.4-1.0%
ScoreClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicAgentic index37.5→36-4.0%
ScoreClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index50.7→50.8+0.2%
ScoreClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicAgentic index56.2→56.5+0.5%
ScoreClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicIntelligence index45.1→44.8-0.7%
ScoreClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicAgentic index47→46.2-1.7%
ScoreClaude Opus 5 (Adaptive Reasoning, Xhigh Effort)AnthropicAgentic index55.5→55.6+0.2%
ScoreClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index30.5→30.1-1.3%
ScoreClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort)AnthropicAgentic index33.1→31.8-3.9%
ScoreClaude Sonnet 5 (Adaptive Reasoning, High Effort)AnthropicIntelligence index32→31.7-0.9%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Low Effort)AnthropicIntelligence index24.7→24.3-1.6%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index38.4→38.2-0.5%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicAgentic index44.3→43.6-1.6%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Medium Effort)AnthropicIntelligence index28.4→28.1-1.1%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)AnthropicIntelligence index34.7→34.4-0.9%
ScoreClaude Sonnet 5 (Non-reasoning, High Effort)AnthropicIntelligence index28.9→23.2-20%
ScoreCommand A+CohereIntelligence index13.9→13.1-5.8%
ScoreCommand A+CohereAgentic index3.6→1-72%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekIntelligence index15.4→14.8-3.9%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekAgentic index8.9→6.8-24%
ScoreDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekIntelligence index24.8→26+4.8%
ScoreDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekAgentic index27.9→Metric withdrawn
ScoreDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekIntelligence index24.6→24.2-1.6%
ScoreDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekAgentic index23.7→22.2-6.3%
ScoreDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekIntelligence index34.5→34.3-0.6%
ScoreDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekAgentic index41.7→41-1.7%
ScoreDeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeekIntelligence index35→34.8-0.6%
ScoreDeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeekAgentic index48.1→47.5-1.2%
ScoreDeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeekIntelligence index30.9→30.4-1.6%
ScoreDeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeekAgentic index27.7→26.3-5.1%
ScoreDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekIntelligence index36.3→36-0.8%
ScoreDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekAgentic index42.3→41.3-2.4%
ScoreDevstral 2MistralIntelligence index9.4→8.6-8.5%
ScoreDevstral 2MistralAgentic index4.9→2.3-53%
ScoreDevstral Small 2MistralIntelligence index8.4→7.5-11%
ScoreDevstral Small 2MistralAgentic index4.8→2-58%
ScoreGemini 2.5 ProGoogleIntelligence index16.7→16.1-3.6%
ScoreGemini 2.5 ProGoogleAgentic index3.5→1.6-54%
ScoreGemini 3.1 Flash-LiteGoogleIntelligence index16→15.6-2.5%
ScoreGemini 3.1 Flash-LiteGoogleAgentic index3.2→1.6-50%
ScoreGemini 3.1 Pro PreviewGoogleIntelligence index30.4→29.7-2.3%
ScoreGemini 3.1 Pro PreviewGoogleAgentic index10.3→8.2-20%
ScoreGemini 3.5 Flash (high)GoogleIntelligence index33→32.6-1.2%
ScoreGemini 3.5 Flash (high)GoogleAgentic index27.3→26-4.8%
ScoreGemini 3.5 Flash-LiteGoogleIntelligence index22.7→22.2-2.2%
ScoreGemini 3.5 Flash-LiteGoogleAgentic index15.9→14.3-10%
ScoreGemini 3.6 Flash (high)GoogleIntelligence index34.3→34-0.9%
ScoreGemini 3.6 Flash (high)GoogleAgentic index30.2→29-4.0%
ScoreGemini 3.7 Flash (high)GoogleIntelligence index39.4→39.1-0.8%
ScoreGemini 3.7 Flash (high)GoogleAgentic index36.4→35.2-3.3%
ScoreGemini 3.8 Flash (high)GoogleIntelligence index41.2→40.9-0.7%
ScoreGemini 3.8 Flash (high)GoogleAgentic index41.1→40.2-2.2%
ScoreGemini 3.8 Flash (low)GoogleIntelligence index33.8→33.5-0.9%
ScoreGemini 3.8 Flash (low)GoogleAgentic index33.7→32.6-3.3%
ScoreGemini 3.8 Flash (medium)GoogleIntelligence index40→39.8-0.5%
ScoreGemini 3.8 Flash (medium)GoogleAgentic index40.5→39.7-2.0%
ScoreGemma 4 31B (Reasoning)GoogleIntelligence index15.4→19+23%
ScoreGemma 4 31B (Reasoning)GoogleAgentic index6.7→4.2-37%
ScoreGLM 5.3 FlashZ AIIntelligence index41.9→41.8-0.2%
ScoreGLM 5.3 FlashZ AIAgentic index51.2→50.9-0.6%
ScoreGLM-5.1 (Reasoning)Z AIIntelligence index26.4→26.1-1.1%
ScoreGLM-5.1 (Reasoning)Z AIAgentic index25.2→23.9-5.2%
ScoreGLM-5.2 (max)Z AIIntelligence index34→33.7-0.9%
ScoreGLM-5.2 (max)Z AIAgentic index39.4→38.4-2.5%
ScoreGLM-5.3 (max)Z AIIntelligence index44.9→44.8-0.2%
ScoreGLM-5.3 (max)Z AIAgentic index53.4→53.1-0.6%
ScoreGPT-5 mini (high)OpenAIIntelligence index17.4→16.8-3.4%
ScoreGPT-5 mini (high)OpenAIAgentic index8.9→6.8-24%
ScoreGPT-5.4 mini (xhigh)OpenAIIntelligence index24.6→24.1-2.0%
ScoreGPT-5.4 mini (xhigh)OpenAIAgentic index19.7→17.9-9.1%
ScoreGPT-5.4 nano (xhigh)OpenAIIntelligence index21.2→20.7-2.4%
ScoreGPT-5.4 nano (xhigh)OpenAIAgentic index17.7→16-9.6%
ScoreGPT-5.5 (high)OpenAIIntelligence index37.3→37-0.8%
ScoreGPT-5.5 (high)OpenAIAgentic index35.4→34.3-3.1%
ScoreGPT-5.5 (medium)OpenAIIntelligence index34.2→33.8-1.2%
ScoreGPT-5.5 (medium)OpenAIAgentic index30.6→29.5-3.6%
ScoreGPT-5.5 (xhigh)OpenAIIntelligence index38.6→38.4-0.5%
ScoreGPT-5.5 (xhigh)OpenAIAgentic index37.3→36.4-2.4%
ScoreGPT-5.5 Instant (June 2026)OpenAIIntelligence index26.8→26-3.0%
ScoreGPT-5.5 Instant (June 2026)OpenAIAgentic index19.2→16.6-14%
ScoreGPT-5.6 Luna (high)OpenAIIntelligence index32.4→32.1-0.9%
ScoreGPT-5.6 Luna (high)OpenAIAgentic index35.6→34.6-2.8%
ScoreGPT-5.6 Luna (low)OpenAIIntelligence index21.5→21-2.3%
ScoreGPT-5.6 Luna (low)OpenAIAgentic index17.9→16.1-10%
ScoreGPT-5.6 Luna (max)OpenAIIntelligence index37.5→37.3-0.5%
ScoreGPT-5.6 Luna (max)OpenAIAgentic index42.7→42.1-1.4%
ScoreGPT-5.6 Luna (medium)OpenAIIntelligence index25.5→25-2.0%
ScoreGPT-5.6 Luna (medium)OpenAIAgentic index25.4→23.9-5.9%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIIntelligence index16.1→15.5-3.7%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIAgentic index15.2→13.4-12%
ScoreGPT-5.6 Luna (xhigh)OpenAIIntelligence index34.8→34.6-0.6%
ScoreGPT-5.6 Luna (xhigh)OpenAIAgentic index39.5→38.7-2.0%
ScoreGPT-5.6 Sol (high)OpenAIIntelligence index42.5→42.3-0.5%
ScoreGPT-5.6 Sol (high)OpenAIAgentic index44.7→44.2-1.1%
ScoreGPT-5.6 Sol (low)OpenAIIntelligence index33.8→33.5-0.9%
ScoreGPT-5.6 Sol (low)OpenAIAgentic index32.6→31.6-3.1%
ScoreGPT-5.6 Sol (max)OpenAIIntelligence index47.1→47-0.2%
ScoreGPT-5.6 Sol (max)OpenAIAgentic index50.5→50.2-0.6%
ScoreGPT-5.6 Sol (medium)OpenAIIntelligence index39.5→39.2-0.8%
ScoreGPT-5.6 Sol (medium)OpenAIAgentic index40.5→39.7-2.0%
ScoreGPT-5.6 Sol (Non-reasoning)OpenAIAgentic index29.4→28.1-4.4%
ScoreGPT-5.6 Sol (xhigh)OpenAIIntelligence index44.1→44-0.2%
ScoreGPT-5.6 Sol (xhigh)OpenAIAgentic index47.8→47.4-0.8%
ScoreGPT-5.6 Terra (high)OpenAIIntelligence index34.5→34.2-0.9%
ScoreGPT-5.6 Terra (high)OpenAIAgentic index37.6→36.7-2.4%
ScoreGPT-5.6 Terra (low)OpenAIIntelligence index27.9→27.5-1.4%
ScoreGPT-5.6 Terra (low)OpenAIAgentic index27.3→25.9-5.1%
ScoreGPT-5.6 Terra (max)OpenAIIntelligence index42.3→42.1-0.5%
ScoreGPT-5.6 Terra (max)OpenAIAgentic index43.7→43.2-1.1%
ScoreGPT-5.6 Terra (medium)OpenAIIntelligence index30.4→30.1-1.0%
ScoreGPT-5.6 Terra (medium)OpenAIAgentic index31.4→30.4-3.2%
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIIntelligence index21.2→20.8-1.9%
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIAgentic index24.9→23.5-5.6%
ScoreGPT-5.6 Terra (xhigh)OpenAIIntelligence index38.2→38-0.5%
ScoreGPT-5.6 Terra (xhigh)OpenAIAgentic index42.2→41.4-1.9%
ScoreGPT-6 Astra (high)OpenAIIntelligence index51→50.9-0.2%
ScoreGPT-6 Astra (high)OpenAIAgentic index48.7→48.2-1.0%
ScoreGPT-6 Astra (low)OpenAIIntelligence index46→45.8-0.4%
ScoreGPT-6 Astra (low)OpenAIAgentic index39.5→38.8-1.8%
ScoreGPT-6 Astra (max)OpenAIIntelligence index52.8→52.7-0.2%
ScoreGPT-6 Astra (max)OpenAIAgentic index51.5→51-1.0%
ScoreGPT-6 Astra (medium)OpenAIIntelligence index49.7→49.6-0.2%
ScoreGPT-6 Astra (medium)OpenAIAgentic index46.3→46-0.6%
ScoreGPT-6 Astra (xhigh)OpenAIIntelligence index52.5→52.4-0.2%
ScoreGPT-6 Astra (xhigh)OpenAIAgentic index50.6→50.2-0.8%
Scoregpt-oss-120b (high)OpenAIIntelligence index12.3→11.6-5.7%
Scoregpt-oss-120b (high)OpenAIAgentic index6.2→3.7-40%
Scoregpt-oss-20b (high)OpenAIAgentic index1.4→1.2-14%
ScoreGranite 4.2 8BIBMIntelligence index11.8→11.1-5.9%
ScoreGranite 4.2 8BIBMAgentic index3.7→1.3-65%
ScoreGrok 4.3 (high)SpaceXAIIntelligence index25.4→24.9-2.0%
ScoreGrok 4.3 (high)SpaceXAIAgentic index17.2→15.5-9.9%
ScoreGrok 4.3 (Non-reasoning)SpaceXAIIntelligence index14.5→14-3.4%
ScoreGrok 4.3 (Non-reasoning)SpaceXAIAgentic index15.8→14-11%
ScoreGrok 4.5 (high)SpaceXAIIntelligence index39.1→38.8-0.8%
ScoreGrok 4.5 (high)SpaceXAIAgentic index42.1→41.2-2.1%
ScoreGrok 4.6 (high)SpaceXAIIntelligence index44.4→44.3-0.2%
ScoreGrok 4.6 (high)SpaceXAIAgentic index53.4→53-0.7%
ScoreGrok 4.6 (low)SpaceXAIIntelligence index35.4→35.1-0.8%
ScoreGrok 4.6 (low)SpaceXAIAgentic index42.2→41.2-2.4%
ScoreGrok 4.6 (medium)SpaceXAIIntelligence index43→42.8-0.5%
ScoreGrok 4.6 (medium)SpaceXAIAgentic index51.2→50.6-1.2%
ScoreGrok 4.6 (xhigh)SpaceXAIIntelligence index44.3→44.2-0.2%
ScoreGrok 4.6 (xhigh)SpaceXAIAgentic index52.7→52.5-0.4%
ScoreHy3TencentIntelligence index25.8→25.3-1.9%
ScoreHy3TencentAgentic index25.6→24.1-5.9%
ScoreHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingAgentic index2.7→0.9-67%
ScoreInkling (xhigh)Thinking MachinesIntelligence index25.5→25-2.0%
ScoreInkling (xhigh)Thinking MachinesAgentic index24.3→22.5-7.4%
ScoreInkling SmallThinking MachinesIntelligence index26.1→27.8+6.5%
ScoreInkling SmallThinking MachinesAgentic index25→Metric withdrawn
ScoreK2 Horizon 3.7BInstitute of Foundation ModelsIntelligence index16.2→15.6-3.7%
ScoreK2 Horizon 3.7BInstitute of Foundation ModelsAgentic index18.4→16.5-10%
ScoreK2 Horizon 375B A23BInstitute of Foundation ModelsIntelligence index30.8→30.5-1.0%
ScoreK2 Horizon 375B A23BInstitute of Foundation ModelsAgentic index40.9→40-2.2%
ScoreK2 Horizon 7BInstitute of Foundation ModelsIntelligence index21→20.6-1.9%
ScoreK2 Horizon 7BInstitute of Foundation ModelsAgentic index27.6→26.1-5.4%
ScoreK2 Horizon MoVA 36B A4BInstitute of Foundation ModelsIntelligence index25.7→25.3-1.6%
ScoreK2 Horizon MoVA 36B A4BInstitute of Foundation ModelsAgentic index34.7→33.5-3.5%
ScoreKimi K2.6KimiIntelligence index27.5→27-1.8%
ScoreKimi K2.6KimiAgentic index22.1→20.5-7.2%
ScoreKimi K2.7 CodeKimiIntelligence index26.3→25.8-1.9%
ScoreKimi K2.7 CodeKimiAgentic index22.5→21-6.7%
ScoreKimi K3 (low)KimiIntelligence index30.5→34.5+13%
ScoreKimi K3 (low)KimiAgentic index30.7→Metric withdrawn
ScoreKimi K3 (max)KimiIntelligence index43.8→43.6-0.5%
ScoreKimi K3 (max)KimiAgentic index50.6→50-1.2%
ScoreLing 3.0 FlashInclusionAIIntelligence index20.6→24.9+21%
ScoreLing 3.0 FlashInclusionAIAgentic index21→Metric withdrawn
ScoreLing 3.0 TinyInclusionAIIntelligence index11.9→15.3+29%
ScoreLing 3.0 TinyInclusionAIAgentic index7.1→Metric withdrawn
ScoreLing-3.0-flash-FinInclusionAIIntelligence index23→22.6-1.7%
ScoreLing-3.0-flash-FinInclusionAIAgentic index29.3→27.9-4.8%
ScoreLing-3.0-flash-VLInclusionAIIntelligence index25→24.6-1.6%
ScoreLing-3.0-flash-VLInclusionAIAgentic index30→28.7-4.3%
ScoreLlama 4 MaverickMetaIntelligence index9.3→10+7.5%
ScoreLlama 4 ScoutMetaIntelligence index6.5→8.1+25%
ScoreLongCat 2.0LongCatIntelligence index19.7→19.1-3.0%
ScoreLongCat 2.0LongCatAgentic index15.9→14-12%
ScoreMercury 2InceptionIntelligence index11.5→13.8+20%
ScoreMercury 2InceptionAgentic index4→1.6-60%
ScoreMiMo-V2.5XiaomiIntelligence index22.3→25.2+13%
ScoreMiMo-V2.5XiaomiAgentic index17.4→15.8-9.2%
ScoreMiMo-V2.5-ProXiaomiIntelligence index26.4→26-1.5%
ScoreMiMo-V2.5-ProXiaomiAgentic index22.7→21.3-6.2%
ScoreMiniCPM5-2BOpenBMBIntelligence index13.1→12.5-4.6%
ScoreMiniCPM5-2BOpenBMBAgentic index9→6.7-26%
ScoreMiniMax-M2.7MiniMaxIntelligence index23.2→22.8-1.7%
ScoreMiniMax-M2.7MiniMaxAgentic index16.8→15.3-8.9%
ScoreMiniMax-M3MiniMaxIntelligence index29.6→29.2-1.4%
ScoreMiniMax-M3MiniMaxAgentic index30.8→29.5-4.2%
ScoreMistral Large 3MistralIntelligence index9.7→9.3-4.1%
ScoreMistral Large 3MistralAgentic index2.4→1-58%
ScoreMistral Medium 3.1MistralIntelligence index9.5→9.2-3.2%
ScoreMistral Medium 3.1MistralAgentic index3.1→1.9-39%
ScoreMistral Medium 3.5MistralIntelligence index14.9→14.2-4.7%
ScoreMistral Medium 3.5MistralAgentic index9.4→7-26%
ScoreMistral Small 3.1MistralIntelligence index7.4→7.1-4.1%
ScoreMistral Small 3.1MistralAgentic index2.1→1.2-43%
ScoreMistral Small 3.2MistralIntelligence index7→8.2+17%
ScoreMistral Small 4 (Reasoning)MistralIntelligence index11.5→11.3-1.7%
ScoreMistral Small 4 (Reasoning)MistralAgentic index1.4→0.8-43%
ScoreMuse Glimmer (high)MetaIntelligence index18.1→17.5-3.3%
ScoreMuse Glimmer (high)MetaAgentic index10.5→8.5-19%
ScoreMuse Spark 1.1 (xhigh)MetaIntelligence index34.3→33.7-1.7%
ScoreMuse Spark 1.1 (xhigh)MetaAgentic index27.5→25.8-6.2%
ScoreMuse Spark 1.2 (xhigh)MetaIntelligence index39.8→39.6-0.5%
ScoreMuse Spark 1.2 (xhigh)MetaAgentic index44→43.2-1.8%
ScoreMuse Spark 1.3 (max)MetaIntelligence index48.2→48.1-0.2%
ScoreMuse Spark 1.3 (max)MetaAgentic index55.7→55.5-0.4%
ScoreMuse Spark 1.3 (xhigh)MetaIntelligence index45.2→45.1-0.2%
ScoreMuse Spark 1.3 (xhigh)MetaAgentic index51.8→51.5-0.6%
ScoreNemotron 3 Super 120B A12B (Reasoning)NVIDIAIntelligence index13.6→12.8-5.9%
ScoreNemotron 3 Super 120B A12B (Reasoning)NVIDIAAgentic index4.1→1.7-59%
ScoreNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAIntelligence index23.4→22.9-2.1%
ScoreNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAAgentic index21.7→20.1-7.4%
ScoreNemotron 3.5 LightningNVIDIAIntelligence index13.6→12.9-5.1%
ScoreNemotron 3.5 LightningNVIDIAAgentic index6.1→3.5-43%
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIAgentic indexFirst measured→23.7
ScoreQuasar 438B (max, based on GLM-5.2)Multiverse ComputingIntelligence index27.1→26.7-1.5%
ScoreQuasar 438B (max, based on GLM-5.2)Multiverse ComputingAgentic index32.7→31.4-4.0%
ScoreQwen3 14B (Reasoning)AlibabaIntelligence index6.4→8.2+28%
ScoreQwen3 32B (Reasoning)AlibabaIntelligence index7.2→8.6+19%
ScoreQwen3 8B (Reasoning)AlibabaIntelligence index5.2→7.3+40%
ScoreQwen3 Coder NextAlibabaIntelligence index10.1→9.2-8.9%
ScoreQwen3 Coder NextAlibabaAgentic index3.6→0.9-75%
ScoreQwen3.5 122B A10B (Reasoning)AlibabaIntelligence index16.2→15.6-3.7%
ScoreQwen3.5 122B A10B (Reasoning)AlibabaAgentic index9.6→7.6-21%
ScoreQwen3.5 397B A17B (Reasoning)AlibabaIntelligence index19.1→18.4-3.7%
ScoreQwen3.5 397B A17B (Reasoning)AlibabaAgentic index10.6→8.3-22%
ScoreQwen3.6 27B (Reasoning)AlibabaIntelligence index21.9→21.4-2.3%
ScoreQwen3.6 27B (Reasoning)AlibabaAgentic index20.1→18.5-8.0%
ScoreQwen3.6 35B A3B (Reasoning)AlibabaIntelligence index18.8→18.2-3.2%
ScoreQwen3.6 35B A3B (Reasoning)AlibabaAgentic index15→13.1-13%
ScoreQwen3.7 MaxAlibabaIntelligence index29.9→29.5-1.3%
ScoreQwen3.7 MaxAlibabaAgentic index23.9→22.5-5.9%
ScoreQwen3.7 PlusAlibabaIntelligence index25.8→25.2-2.3%
ScoreQwen3.7 PlusAlibabaAgentic index19.7→17.5-11%
ScoreQwen3.8 2.4T A95BAlibabaIntelligence index40→39.9-0.3%
ScoreQwen3.8 2.4T A95BAlibabaAgentic index50.4→50.1-0.6%
ScoreQwen3.8 27B (low)AlibabaIntelligence index26.5→26.2-1.1%
ScoreQwen3.8 27B (low)AlibabaAgentic index39.9→38.9-2.5%
ScoreQwen3.8 27B (medium)AlibabaIntelligence index27.8→27.6-0.7%
ScoreQwen3.8 27B (medium)AlibabaAgentic index45.3→44.4-2.0%
ScoreQwen3.8 27B (Non-reasoning)AlibabaAgentic indexFirst measured→22.4
ScoreQwen3.8 27B (xhigh)AlibabaIntelligence index33.9→33.7-0.6%
ScoreQwen3.8 27B (xhigh)AlibabaAgentic index46.5→45.8-1.5%
ScoreQwen3.8 MaxAlibabaIntelligence index40.3→40.2-0.2%
ScoreQwen3.8 MaxAlibabaAgentic index49.6→49.1-1.0%
ScoreQwen3.8 Max (0902)AlibabaAgentic index56.1→56-0.2%
ScoreQwen3.8-Flash-NextAlibabaIntelligence index39.9→39.8-0.3%
ScoreQwen3.8-Flash-NextAlibabaAgentic index53.9→53.6-0.6%
ScoreRing-2.6-1TInclusionAIIntelligence index17.3→16.6-4.0%
ScoreRing-2.6-1TInclusionAIAgentic index12.9→10.7-17%
ScoreTrinity Large ThinkingArcee AIIntelligence index10.9→10.8-0.9%
ScoreTrinity Large ThinkingArcee AIAgentic index1.2→1-17%

09-17

Sep 17, 20267 changes

AddedLing-3.0-flash-FinInclusionAIIntelligence index 23
PriceHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingInput price$0.040→Metric withdrawn
PriceHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingOutput price$0.140→Metric withdrawn
PriceLlama 3.3 Instruct 70BMetaInput price$0.660→$0.710+7.6%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIIntelligence index16.8→16.1-4.2%
ScoreMistral Medium 3.1MistralIntelligence index9.9→9.5-4.0%
ScoreNorth Mini CodeCohereIntelligence index12.8→9.9-23%

09-16

Sep 16, 202611 changes

AddedQwen3.8 Max (0902)AlibabaIntelligence index 45.4
PriceClaude 4 Sonnet (Non-reasoning)AnthropicInput price$3.00→Metric withdrawn
PriceClaude 4 Sonnet (Non-reasoning)AnthropicOutput price$15.00→Metric withdrawn
PriceClaude 4 Sonnet (Reasoning)AnthropicInput price$3.00→Metric withdrawn
PriceClaude 4 Sonnet (Reasoning)AnthropicOutput price$15.00→Metric withdrawn
PriceNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIInput price$0.500→Metric withdrawn
PriceNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIOutput price$2.50→Metric withdrawn
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIIntelligence index22.3→21.2-4.9%
ScoreKimi K2.6KimiIntelligence index31.3→27.5-12%
ScoreLing 3.0 FlashInclusionAIIntelligence index24.9→20.6-17%
ScoreLing-3.0-flash-VLInclusionAIIntelligence index24.8→25+0.8%

09-15

Sep 15, 20269 changes

AddedClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index 42
AddedK2 Horizon 0.9BMBZUAI Institute of Foundation ModelsIntelligence index 3
AddedK2 Horizon 3.7BMBZUAI Institute of Foundation ModelsIntelligence index 16.2
AddedK2 Horizon 7BMBZUAI Institute of Foundation ModelsIntelligence index 21
AddedK2 Horizon MoVA 36B A4BMBZUAI Institute of Foundation ModelsIntelligence index 25.7
PriceHy3TencentOutput price$0.550→$0.580+5.5%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAOutput price$2.60→$2.40-7.7%
ScoreGLM-5.2 (max)Z AIIntelligence index38.6→34-12%
ScoreKimi K3 (low)KimiIntelligence index34.5→30.5-12%

09-13

Sep 13, 20263 changes

RemovedClaude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic
RemovedGPT-6 Astra (Non-reasoning)OpenAI
ScoreK2 Horizon 375B A23BMBZUAI Institute of Foundation ModelsIntelligence index30.7→30.8+0.3%

09-12

Sep 12, 20266 changes

AddedAgnes 3.0 FlashSapiens AIIntelligence index 35.5
ScoreAgnes 2.5 Pro AlphaSapiens AIIntelligence index26.8→27.8+3.7%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)AnthropicIntelligence indexFirst measured→34.7
ScoreGPT-5.6 Luna (low)OpenAIIntelligence index21.8→21.5-1.4%
ScoreGPT-5.6 Luna (medium)OpenAIIntelligence index25.8→25.5-1.2%
ScoreK2 Horizon 375B A23BMBZUAI Institute of Foundation ModelsIntelligence index33.9→30.7-9.4%

09-11

Sep 11, 202618 changes

AddedDeepSeek V4.1 Flash (Reasoning, Max Effort)DeepSeekIntelligence index 39.5
AddedLing-3.0-flash-VLInclusionAIIntelligence index 24.8
PriceGPT-6 Astra (Non-reasoning)OpenAIInput price$10.00→Metric withdrawn
PriceGPT-6 Astra (Non-reasoning)OpenAIOutput price$50.00→Metric withdrawn
PriceLongCat 2.0LongCatInput price$0.750→$0.300-60%
PriceLongCat 2.0LongCatOutput price$2.95→$1.20-59%
ScoreClaude Sonnet 5 (Adaptive Reasoning, High Effort)AnthropicIntelligence indexFirst measured→32
ScoreGLM-5.1 (Reasoning)Z AIIntelligence index27.4→26.4-3.6%
ScoreGPT-5.6 Luna (high)OpenAIIntelligence index32.9→32.4-1.5%
ScoreGPT-5.6 Terra (medium)OpenAIIntelligence index32.8→30.4-7.3%
ScoreGranite 4.2 8BIBMIntelligence index12.4→11.8-4.8%
ScoreMiniCPM5-2BOpenBMBIntelligence index14.3→13.1-8.4%
ScoreQwen3.6 35B A3B (Reasoning)AlibabaIntelligence index22.3→18.8-16%
ScoreQwen3.8 27B (low)AlibabaIntelligence index29.2→26.5-9.2%
ScoreQwen3.8 27B (medium)AlibabaIntelligence index30.7→27.8-9.4%
ScoreQwen3.8 27B (medium)AlibabaAgentic indexFirst measured→45.3
ScoreQwen3.8-Flash-NextAlibabaIntelligence index42.2→39.9-5.5%
ScoreQwen3.8-Flash-NextAlibabaAgentic indexFirst measured→53.9

09-09

Sep 9, 2026753 changes

AddedMuse Spark 1.3 (max)MetaIntelligence index 48.2
RemovedMuse Spark 1.3 (max)Meta
PriceDeepSeek V3 (Dec '24)DeepSeekInput price$0.360→$0.320-11%
PriceDeepSeek V3 0324DeepSeekInput price$0.270→$0.840+211%
PriceDeepSeek V3 0324DeepSeekOutput price$1.12→$1.18+5.4%
PriceHermes 4 - Llama-3.1 70B (Non-reasoning)Nous ResearchInput price$0.130→Metric withdrawn
PriceHermes 4 - Llama-3.1 70B (Non-reasoning)Nous ResearchOutput price$0.400→Metric withdrawn
PriceHermes 4 - Llama-3.1 70B (Reasoning)Nous ResearchInput price$0.130→Metric withdrawn
PriceHermes 4 - Llama-3.1 70B (Reasoning)Nous ResearchOutput price$0.400→Metric withdrawn
PriceLlama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIAInput price$0.600→Metric withdrawn
PriceLlama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIAOutput price$1.80→Metric withdrawn
PriceNemotron 3 Nano Omni 30B A3B ReasoningNVIDIAInput price$0.070→$0.090+29%
PriceNemotron 3 Nano Omni 30B A3B ReasoningNVIDIAOutput price$0.300→$0.360+20%
PriceNemotron 3 Super 120B A12B (Reasoning)NVIDIAInput price$0.200→$0.190-5.0%
PriceNemotron 3 Super 120B A12B (Reasoning)NVIDIAOutput price$0.800→$0.650-19%
ScoreA.X-K2SK TelecomIntelligence index26.8→22.7-15%
ScoreAgnes 2.5 Pro AlphaSapiens AIIntelligence index31.2→26.8-14%
ScoreAgnes 2.5 Pro AlphaSapiens AIAgentic index29.5→29.4-0.3%
ScoreAgnes 2.5 Pro BetaSapiens AIIntelligence index39.5→35.2-11%
ScoreApertus 70B InstructSwiss AI InitiativeIntelligence index1→5.1+410%
ScoreApertus 8B InstructSwiss AI InitiativeIntelligence index1→4.8+380%
ScoreApodex 1.1ApodexIntelligence index34.9→30.4-13%
ScoreApriel-v1.5-15B-ThinkerServiceNowIntelligence index15→13.8-8.0%
ScoreApriel-v1.6-15B-ThinkerServiceNowIntelligence index14.3→13.4-6.3%
ScoreArctic InstructSnowflakeIntelligence index1→5.4+440%
ScoreCeleris-1CelerisIntelligence index7.6→6.3-17%
ScoreClaude 2.0AnthropicIntelligence index1→5.5+450%
ScoreClaude 2.1AnthropicIntelligence index1→5.6+460%
ScoreClaude 3 HaikuAnthropicIntelligence index1→5.6+460%
ScoreClaude 3 OpusAnthropicIntelligence index6→8.7+45%
ScoreClaude 3 SonnetAnthropicIntelligence index1→5.9+490%
ScoreClaude 3.5 HaikuAnthropicIntelligence index6.5→8.9+37%
ScoreClaude 3.5 Sonnet (June '24)AnthropicIntelligence index2.7→7.2+167%
ScoreClaude 3.5 Sonnet (Oct '24)AnthropicIntelligence index4.2→7.9+88%
ScoreClaude 3.7 Sonnet (Non-reasoning)AnthropicIntelligence index17.2→15.3-11%
ScoreClaude 3.7 Sonnet (Reasoning)AnthropicIntelligence index20.5→17.7-14%
ScoreClaude 4 Opus (Non-reasoning)AnthropicIntelligence index19.1→16.6-13%
ScoreClaude 4 Opus (Reasoning)AnthropicIntelligence index24.3→20.6-15%
ScoreClaude 4 Sonnet (Non-reasoning)AnthropicIntelligence index19.1→16.6-13%
ScoreClaude 4 Sonnet (Reasoning)AnthropicIntelligence index22.2→18.9-15%
ScoreClaude 4.1 Opus (Non-reasoning)AnthropicIntelligence index21.7→18.6-14%
ScoreClaude 4.1 Opus (Reasoning)AnthropicIntelligence index26.9→22.8-15%
ScoreClaude 4.5 Haiku (Non-reasoning)AnthropicIntelligence index17.4→15.4-11%
ScoreClaude 4.5 Haiku (Reasoning)AnthropicIntelligence index22.5→17.6-22%
ScoreClaude 4.5 Haiku (Reasoning)AnthropicAgentic index10.4→10.3-1.0%
ScoreClaude 4.5 Sonnet (Non-reasoning)AnthropicIntelligence index22.7→19.3-15%
ScoreClaude 4.5 Sonnet (Reasoning)AnthropicIntelligence index27.3→21.2-22%
ScoreClaude 4.5 Sonnet (Reasoning)AnthropicAgentic index17.6→17.5-0.6%
ScoreClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)AnthropicIntelligence index53.2→49.7-6.6%
ScoreClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)AnthropicAgentic index51.2→51-0.4%
ScoreClaude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)AnthropicIntelligence index54.5→51.2-6.1%
ScoreClaude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)AnthropicAgentic index53.5→53.3-0.4%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)AnthropicIntelligence index50.9→47-7.7%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)AnthropicAgentic index48→47.8-0.4%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)AnthropicIntelligence index56.8→53.4-6.0%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)AnthropicAgentic index58.2→58-0.3%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)AnthropicIntelligence index52.8→49.1-7.0%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)AnthropicAgentic index50.9→50.6-0.6%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)AnthropicIntelligence index56.2→53.2-5.3%
ScoreClaude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)AnthropicAgentic index57.4→57.1-0.5%
ScoreClaude InstantAnthropicIntelligence index1→5+400%
ScoreClaude Opus 4.5 (Non-reasoning)AnthropicIntelligence index27.8→23.7-15%
ScoreClaude Opus 4.5 (Reasoning)AnthropicIntelligence index33.6→29.1-13%
ScoreClaude Opus 4.6 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index36.4→31.9-12%
ScoreClaude Opus 4.6 (Non-reasoning, High Effort)AnthropicIntelligence index30.8→26.4-14%
ScoreClaude Opus 4.7 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index44.3→40.7-8.1%
ScoreClaude Opus 4.7 (Adaptive Reasoning, Max Effort)AnthropicAgentic index39.7→39.5-0.5%
ScoreClaude Opus 4.7 (Non-reasoning, High Effort)AnthropicIntelligence index35.4→30.9-13%
ScoreClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index47.8→42-12%
ScoreClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicAgentic index42.8→42.6-0.5%
ScoreClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicIntelligence index52→48.2-7.3%
ScoreClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicAgentic index52.9→52.7-0.4%
ScoreClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicIntelligence index43.8→39.8-9.1%
ScoreClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicAgentic index37.7→37.5-0.5%
ScoreClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index54.1→50.7-6.3%
ScoreClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicAgentic index56.4→56.2-0.4%
ScoreClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicIntelligence index49.5→45.1-8.9%
ScoreClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicAgentic index47.2→47-0.4%
ScoreClaude Opus 5 (Adaptive Reasoning, Xhigh Effort)AnthropicIntelligence index53.4→49.7-6.9%
ScoreClaude Opus 5 (Adaptive Reasoning, Xhigh Effort)AnthropicAgentic index55.7→55.5-0.4%
ScoreClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index38.5→30.5-21%
ScoreClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort)AnthropicAgentic index33.3→33.1-0.6%
ScoreClaude Sonnet 4.6 (Non-reasoning, High Effort)AnthropicIntelligence index29→24.7-15%
ScoreClaude Sonnet 4.6 (Non-reasoning, Low Effort)AnthropicIntelligence index27.4→23.3-15%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Low Effort)AnthropicIntelligence indexFirst measured→24.7
ScoreClaude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index45.1→38.4-15%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicAgentic index44.5→44.3-0.4%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Medium Effort)AnthropicIntelligence indexFirst measured→28.4
ScoreClaude Sonnet 5 (Non-reasoning, High Effort)AnthropicIntelligence index33.5→28.9-14%
ScoreCommand ACohereIntelligence index2.1→7+233%
ScoreCommand A+CohereIntelligence index15.6→13.9-11%
ScoreCommand A+CohereAgentic index3.7→3.6-2.7%
ScoreCommand-R (Mar '24)CohereIntelligence index1→5+400%
ScoreCommand-R+ (Apr '24)CohereIntelligence index1→5.3+430%
ScoreDBRX InstructDatabricksIntelligence index1→5.3+430%
ScoreDeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous ResearchIntelligence index1→5.1+410%
ScoreDeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous ResearchIntelligence index1→6.1+510%
ScoreDeepSeek Coder V2 Lite InstructDeepSeekIntelligence index1→5.3+430%
ScoreDeepSeek LLM 67B Chat (V1)DeepSeekIntelligence index1→5.3+430%
ScoreDeepSeek R1 (Jan '25)DeepSeekIntelligence index12→11.4-5.0%
ScoreDeepSeek R1 0528 (May '25)DeepSeekIntelligence index13.9→13.1-5.8%
ScoreDeepSeek R1 0528 Qwen3 8BDeepSeekIntelligence index4.7→8.1+72%
ScoreDeepSeek R1 Distill Llama 70BDeepSeekIntelligence index4.2→7.9+88%
ScoreDeepSeek R1 Distill Llama 8BDeepSeekIntelligence index1→6.5+550%
ScoreDeepSeek R1 Distill Qwen 1.5BDeepSeekIntelligence index1→5.5+450%
ScoreDeepSeek R1 Distill Qwen 14BDeepSeekIntelligence index4.1→7.8+90%
ScoreDeepSeek R1 Distill Qwen 32BDeepSeekIntelligence index5.3→8.4+58%
ScoreDeepSeek V3 (Dec '24)DeepSeekIntelligence index10.4→8.5-18%
ScoreDeepSeek V3 0324DeepSeekIntelligence index9.2→9.7+5.4%
ScoreDeepSeek V3.1 (Non-reasoning)DeepSeekIntelligence index14.8→13.7-7.4%
ScoreDeepSeek V3.1 (Reasoning)DeepSeekIntelligence index14.5→13.5-6.9%
ScoreDeepSeek V3.1 Terminus (Non-reasoning)DeepSeekIntelligence index15.2→13.9-8.6%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekIntelligence index23.5→15.4-34%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekAgentic index9→8.9-1.1%
ScoreDeepSeek V3.2 (Non-reasoning)DeepSeekIntelligence index18.3→16-13%
ScoreDeepSeek V3.2 (Reasoning)DeepSeekIntelligence index25.3→21.5-15%
ScoreDeepSeek V3.2 Exp (Non-reasoning)DeepSeekIntelligence index15.1→13.9-7.9%
ScoreDeepSeek V3.2 Exp (Reasoning)DeepSeekIntelligence index19→16.6-13%
ScoreDeepSeek V3.2 SpecialeDeepSeekIntelligence index16→14.5-9.4%
ScoreDeepSeek V4 Flash 0420 (Non-reasoning)DeepSeekIntelligence index22.1→18.9-14%
ScoreDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekIntelligence index30.2→24.8-18%
ScoreDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekAgentic index28.1→27.9-0.7%
ScoreDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekIntelligence index30.7→24.6-20%
ScoreDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekAgentic index23.8→23.7-0.4%
ScoreDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekIntelligence index40.8→34.5-15%
ScoreDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekAgentic index41.9→41.7-0.5%
ScoreDeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeekIntelligence index40.7→35-14%
ScoreDeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeekAgentic index48.2→48.1-0.2%
ScoreDeepSeek V4 Pro 0424 (Non-reasoning)DeepSeekIntelligence index24.5→20.8-15%
ScoreDeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeekIntelligence index34.6→30.1-13%
ScoreDeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeekIntelligence index34.5→30.9-10%
ScoreDeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeekAgentic index27.9→27.7-0.7%
ScoreDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekIntelligence index42.1→36.3-14%
ScoreDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekAgentic index42.5→42.3-0.5%
ScoreDeepSeek-Coder-V2DeepSeekIntelligence index1→6+500%
ScoreDeepSeek-V2-ChatDeepSeekIntelligence index1→5.5+450%
ScoreDeepSeek-V2.5DeepSeekIntelligence index1.1→6.6+500%
ScoreDeepSeek-V2.5 (Dec '24)DeepSeekIntelligence index1.2→6.6+450%
ScoreDevstral 2MistralIntelligence index12.7→9.4-26%
ScoreDevstral MediumMistralIntelligence index6.6→9+36%
ScoreDevstral Small (Jul '25)MistralIntelligence index3.6→7.6+111%
ScoreDevstral Small (May '25)MistralIntelligence index6→8.7+45%
ScoreDevstral Small 2MistralIntelligence index11.7→8.4-28%
ScoreDiffusionGemma 26B A4BGoogleIntelligence index7.6→9.5+25%
ScoreDoubao Seed CodeByteDance SeedIntelligence index19.5→16.9-13%
ScoreERNIE 4.5 300B A47BBaiduIntelligence index3.4→7.5+121%
ScoreERNIE 5.0 Thinking PreviewBaiduIntelligence index15.7→14.3-8.9%
ScoreExaone 4.0 1.2B (Non-reasoning)LG AI ResearchIntelligence index1→5.2+420%
ScoreExaone 4.0 1.2B (Reasoning)LG AI ResearchIntelligence index1→5.3+430%
ScoreEXAONE 4.0 32B (Non-reasoning)LG AI ResearchIntelligence index1→6.3+530%
ScoreEXAONE 4.0 32B (Reasoning)LG AI ResearchIntelligence index4.9→8.2+67%
ScoreEXAONE 4.5 33BLG AI ResearchIntelligence index14→13.2-5.7%
ScoreFalcon-H1R-7BTII UAEIntelligence index4.1→7.8+90%
ScoreG9v3-39A5BAI9StarsIntelligence index25.7→21.8-15%
ScoreG9v3-3BAI9StarsIntelligence index10.1→10.8+6.9%
ScoreGemini 1.0 ProGoogleIntelligence index1→5.3+430%
ScoreGemini 1.0 UltraGoogleIntelligence index1→5.8+480%
ScoreGemini 1.5 Flash (May '24)GoogleIntelligence index1→5.9+490%
ScoreGemini 1.5 Flash (Sep '24)GoogleIntelligence index2.4→7.1+196%
ScoreGemini 1.5 Flash-8BGoogleIntelligence index1→6.2+520%
ScoreGemini 1.5 Pro (May '24)GoogleIntelligence index1→6.4+540%
ScoreGemini 1.5 Pro (Sep '24)GoogleIntelligence index4.3→7.9+84%
ScoreGemini 2.0 Flash (experimental)GoogleIntelligence index5→8.2+64%
ScoreGemini 2.0 Flash (Feb '25)GoogleIntelligence index6.5→8.9+37%
ScoreGemini 2.0 Flash Thinking Experimental (Dec '24)GoogleIntelligence index1.1→6.6+500%
ScoreGemini 2.0 Flash Thinking Experimental (Jan '25)GoogleIntelligence index7.4→9.4+27%
ScoreGemini 2.0 Flash-Lite (Feb '25)GoogleIntelligence index3.2→7.4+131%
ScoreGemini 2.0 Flash-Lite (Preview)GoogleIntelligence index3→7.3+143%
ScoreGemini 2.0 Pro Experimental (Feb '25)GoogleIntelligence index6.1→8.7+43%
ScoreGemini 2.5 Flash (Non-reasoning)GoogleIntelligence index8.2→9.9+21%
ScoreGemini 2.5 Flash (Reasoning)GoogleIntelligence index13.9→13.1-5.8%
ScoreGemini 2.5 Flash Preview (Non-reasoning)GoogleIntelligence index5.9→8.7+47%
ScoreGemini 2.5 Flash Preview (Reasoning)GoogleIntelligence index11.5→11.7+1.7%
ScoreGemini 2.5 Flash Preview (Sep '25) (Non-reasoning)GoogleIntelligence index12.7→12.4-2.4%
ScoreGemini 2.5 Flash Preview (Sep '25) (Reasoning)GoogleIntelligence index17.4→15.5-11%
ScoreGemini 2.5 Flash-Lite (Non-reasoning)GoogleIntelligence index1.4→6.7+379%
ScoreGemini 2.5 Flash-Lite (Reasoning)GoogleIntelligence index5.6→8.5+52%
ScoreGemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)GoogleIntelligence index7.3→9.3+27%
ScoreGemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)GoogleIntelligence index9.2→10.4+13%
ScoreGemini 2.5 ProGoogleIntelligence index19→16.7-12%
ScoreGemini 2.5 ProGoogleAgentic index3.6→3.5-2.8%
ScoreGemini 2.5 Pro Preview (Mar' 25)GoogleIntelligence index16.7→15-10%
ScoreGemini 2.5 Pro Preview (May' 25)GoogleIntelligence index16.1→14.5-9.9%
ScoreGemini 3 Flash Preview (Non-reasoning)GoogleIntelligence index20.8→17.9-14%
ScoreGemini 3 Flash Preview (Reasoning)GoogleIntelligence index30.7→26.3-14%
ScoreGemini 3 Pro Preview (high)GoogleIntelligence index32.4→28-14%
ScoreGemini 3 Pro Preview (low)GoogleIntelligence index26.3→22.3-15%
ScoreGemini 3.1 Flash-LiteGoogleIntelligence index19.3→16-17%
ScoreGemini 3.1 Flash-LiteGoogleAgentic index3.3→3.2-3.0%
ScoreGemini 3.1 Pro PreviewGoogleIntelligence index36.7→30.4-17%
ScoreGemini 3.1 Pro PreviewGoogleAgentic index10.4→10.3-1.0%
ScoreGemini 3.5 Flash (high)GoogleIntelligence index39.7→33-17%
ScoreGemini 3.5 Flash (high)GoogleAgentic index27.4→27.3-0.4%
ScoreGemini 3.5 Flash (medium)GoogleIntelligence index38→33.6-12%
ScoreGemini 3.5 Flash (minimal)GoogleIntelligence index28→23.8-15%
ScoreGemini 3.5 Flash-LiteGoogleIntelligence index27.6→22.7-18%
ScoreGemini 3.5 Flash-LiteGoogleAgentic index16.1→15.9-1.2%
ScoreGemini 3.6 Flash (high)GoogleIntelligence index40.3→34.3-15%
ScoreGemini 3.6 Flash (high)GoogleAgentic index30.3→30.2-0.3%
ScoreGemini 3.7 Flash (high)GoogleIntelligence index45.2→39.4-13%
ScoreGemini 3.7 Flash (high)GoogleAgentic index36.6→36.4-0.5%
ScoreGemini 3.7 Flash (low)GoogleIntelligence index41.1→36.9-10%
ScoreGemini 3.7 Flash (medium)GoogleIntelligence index43.4→39.6-8.8%
ScoreGemini 3.8 Flash (high)GoogleIntelligence index47.1→41.2-13%
ScoreGemini 3.8 Flash (high)GoogleAgentic index41.2→41.1-0.2%
ScoreGemini 3.8 Flash (low)GoogleIntelligence index41→33.8-18%
ScoreGemini 3.8 Flash (low)GoogleAgentic index33.8→33.7-0.3%
ScoreGemini 3.8 Flash (medium)GoogleIntelligence index46.8→40-15%
ScoreGemini 3.8 Flash (medium)GoogleAgentic indexFirst measured→40.5
ScoreGemma 3 12B InstructGoogleIntelligence index1→3.8+280%
ScoreGemma 3 1B InstructGoogleIntelligence index1→4.8+380%
ScoreGemma 3 270MGoogleIntelligence index1→5.1+410%
ScoreGemma 3 27B InstructGoogleIntelligence index2→4.9+145%
ScoreGemma 3 4B InstructGoogleIntelligence index1→4.8+380%
ScoreGemma 3n E2B InstructGoogleIntelligence index1→4.8+380%
ScoreGemma 3n E4B InstructGoogleIntelligence index1→4.8+380%
ScoreGemma 3n E4B Instruct Preview (May '25)GoogleIntelligence index1→5.8+480%
ScoreGemma 4 12B (Non-reasoning)GoogleIntelligence index7.3→9.4+29%
ScoreGemma 4 12B (Reasoning)GoogleIntelligence index15.6→14.2-9.0%
ScoreGemma 4 26B A4B (Non-reasoning)GoogleIntelligence index13.9→13.1-5.8%
ScoreGemma 4 26B A4B (Reasoning)GoogleIntelligence index19.2→16.7-13%
ScoreGemma 4 31B (Non-reasoning)GoogleIntelligence index15.2→13.9-8.6%
ScoreGemma 4 31B (Reasoning)GoogleIntelligence index22.2→15.4-31%
ScoreGemma 4 31B (Reasoning)GoogleAgentic index6.8→6.7-1.5%
ScoreGemma 4 E2B (Non-reasoning)GoogleIntelligence index1→6.5+550%
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index4→7.8+95%
ScoreGemma 4 E4B (Non-reasoning)GoogleIntelligence index3.3→7.5+127%
ScoreGemma 4 E4B (Reasoning)GoogleIntelligence index6.4→8.9+39%
ScoreGLM 5V Turbo (Reasoning)Z AIIntelligence index27.6→23.5-15%
ScoreGLM-4.5 (Reasoning)Z AIIntelligence index13.3→12.8-3.8%
ScoreGLM-4.5-AirZ AIIntelligence index10.5→11.1+5.7%
ScoreGLM-4.5V (Non-reasoning)Z AIIntelligence index1.4→6.7+379%
ScoreGLM-4.5V (Reasoning)Z AIIntelligence index3.5→7.6+117%
ScoreGLM-4.6 (Non-reasoning)Z AIIntelligence index16.7→14.9-11%
ScoreGLM-4.6 (Reasoning)Z AIIntelligence index21.7→18.5-15%
ScoreGLM-4.6V (Non-reasoning)Z AIIntelligence index5.2→8.4+62%
ScoreGLM-4.6V (Reasoning)Z AIIntelligence index10.7→11.2+4.7%
ScoreGLM-4.7 (Non-reasoning)Z AIIntelligence index20.1→17.4-13%
ScoreGLM-4.7 (Reasoning)Z AIIntelligence index26.2→22.2-15%
ScoreGLM-4.7-Flash (Non-reasoning)Z AIIntelligence index9.6→10.6+10%
ScoreGLM-4.7-Flash (Reasoning)Z AIIntelligence index16.6→14.9-10%
ScoreGLM-5 (Non-reasoning)Z AIIntelligence index25.6→21.8-15%
ScoreGLM-5 (Reasoning)Z AIIntelligence index32.4→27.9-14%
ScoreGLM-5-TurboZ AIIntelligence index31→26.6-14%
ScoreGLM-5.1 (Non-reasoning)Z AIIntelligence index28.5→24.2-15%
ScoreGLM-5.1 (Reasoning)Z AIIntelligence index31.9→27.4-14%
ScoreGLM-5.1 (Reasoning)Z AIAgentic index25.4→25.2-0.8%
ScoreGLM-5.2 (max)Z AIIntelligence index42.1→38.6-8.3%
ScoreGLM-5.2 (max)Z AIAgentic index39.7→39.4-0.8%
ScoreGLM-5.2 (Non-reasoning)Z AIIntelligence index26.5→22.4-15%
ScoreGLM-5.3 (max)Z AIIntelligence index48.6→44.9-7.6%
ScoreGLM-5.3 (max)Z AIAgentic index53.6→53.4-0.4%
ScoreGLM 5.3 FlashZ AIIntelligence index46.2→41.9-9.3%
ScoreGLM 5.3 FlashZ AIAgentic index51.5→51.2-0.6%
ScoreGPT-3.5 TurboOpenAIIntelligence index1→5.5+450%
ScoreGPT-4OpenAIIntelligence index1.5→6.7+347%
ScoreGPT-4 TurboOpenAIIntelligence index2.3→7+204%
ScoreGPT-4.1OpenAIIntelligence index13.2→12.7-3.8%
ScoreGPT-4.1 miniOpenAIIntelligence index8.8→10.2+16%
ScoreGPT-4.1 nanoOpenAIIntelligence index4.1→7.8+90%
ScoreGPT-4.5 (Preview)OpenAIIntelligence index7.7→9.6+25%
ScoreGPT-4o (Aug '24)OpenAIIntelligence index3.9→7.7+97%
ScoreGPT-4o (ChatGPT)OpenAIIntelligence index2.6→7.2+177%
ScoreGPT-4o (March 2025, chatgpt-4o-latest)OpenAIIntelligence index6.5→9+38%
ScoreGPT-4o (May '24)OpenAIIntelligence index3→7.3+143%
ScoreGPT-4o (Nov '24)OpenAIIntelligence index5.4→8.4+56%
ScoreGPT-4o miniOpenAIIntelligence index1.4→6.7+379%
ScoreGPT-5 (ChatGPT)OpenAIIntelligence index9.4→10.4+11%
ScoreGPT-5 (high)OpenAIIntelligence index27.1→23-15%
ScoreGPT-5 (low)OpenAIIntelligence index24.4→20.8-15%
ScoreGPT-5 (medium)OpenAIIntelligence index26.9→22.9-15%
ScoreGPT-5 (minimal)OpenAIIntelligence index11.1→11.4+2.7%
ScoreGPT-5 Codex (high)OpenAIIntelligence index29.2→24.9-15%
ScoreGPT-5 mini (high)OpenAIIntelligence index18.4→17.4-5.4%
ScoreGPT-5 mini (high)OpenAIAgentic index9→8.9-1.1%
ScoreGPT-5 mini (medium)OpenAIIntelligence index24.2→20.6-15%
ScoreGPT-5 mini (minimal)OpenAIIntelligence index8.4→9.9+18%
ScoreGPT-5 nano (high)OpenAIIntelligence index13.7→13-5.1%
ScoreGPT-5 nano (medium)OpenAIIntelligence index12.9→12.5-3.1%
ScoreGPT-5 nano (minimal)OpenAIIntelligence index2.4→7.1+196%
ScoreGPT-5.1 (high)OpenAIIntelligence index29.1→24.7-15%
ScoreGPT-5.1 (Non-reasoning)OpenAIIntelligence index14.2→13.3-6.3%
ScoreGPT-5.1 Codex (high)OpenAIIntelligence index27.9→23.7-15%
ScoreGPT-5.1 Codex mini (high)OpenAIIntelligence index24→20.4-15%
ScoreGPT-5.2 (medium)OpenAIIntelligence index30.9→26.5-14%
ScoreGPT-5.2 (Non-reasoning)OpenAIIntelligence index19.6→17-13%
ScoreGPT-5.2 (xhigh)OpenAIIntelligence index35→30.4-13%
ScoreGPT-5.2 Codex (xhigh)OpenAIIntelligence index33→28.5-14%
ScoreGPT-5.3 Codex (xhigh)OpenAIIntelligence index36.9→32.5-12%
ScoreGPT-5.4 (low)OpenAIIntelligence index32.1→27.6-14%
ScoreGPT-5.4 (Non-reasoning)OpenAIIntelligence index21.1→18.2-14%
ScoreGPT-5.4 (xhigh)OpenAIIntelligence index42.8→39-8.9%
ScoreGPT-5.4 mini (medium)OpenAIIntelligence index23.2→19.7-15%
ScoreGPT-5.4 mini (Non-Reasoning)OpenAIIntelligence index10.6→11.1+4.7%
ScoreGPT-5.4 mini (xhigh)OpenAIIntelligence index31.9→24.6-23%
ScoreGPT-5.4 mini (xhigh)OpenAIAgentic index19.8→19.7-0.5%
ScoreGPT-5.4 nano (medium)OpenAIIntelligence index23.5→20-15%
ScoreGPT-5.4 nano (Non-Reasoning)OpenAIIntelligence index11.5→11.7+1.7%
ScoreGPT-5.4 nano (xhigh)OpenAIIntelligence index30.9→21.2-31%
ScoreGPT-5.4 nano (xhigh)OpenAIAgentic index17.9→17.7-1.1%
ScoreGPT-5.5 (high)OpenAIIntelligence index44.1→37.3-15%
ScoreGPT-5.5 (high)OpenAIAgentic index35.6→35.4-0.6%
ScoreGPT-5.5 (low)OpenAIIntelligence index35.3→30.7-13%
ScoreGPT-5.5 (medium)OpenAIIntelligence index41.2→34.2-17%
ScoreGPT-5.5 (medium)OpenAIAgentic index30.7→30.6-0.3%
ScoreGPT-5.5 (Non-reasoning)OpenAIIntelligence index27.3→23.2-15%
ScoreGPT-5.5 (xhigh)OpenAIIntelligence index45.6→38.6-15%
ScoreGPT-5.5 (xhigh)OpenAIAgentic index37.5→37.3-0.5%
ScoreGPT-5.5 Instant (June 2026)OpenAIIntelligence index21.6→26.8+24%
ScoreGPT-5.5 Instant (June 2026)OpenAIAgentic index19.4→19.2-1.0%
ScoreGPT-5.5 Instant (May 2026)OpenAIIntelligence index26.7→22.7-15%
ScoreGPT-5.6 Luna (high)OpenAIIntelligence index37.4→32.9-12%
ScoreGPT-5.6 Luna (high)OpenAIAgentic index35.8→35.6-0.6%
ScoreGPT-5.6 Luna (low)OpenAIIntelligence index25.8→21.8-16%
ScoreGPT-5.6 Luna (low)OpenAIAgentic index18→17.9-0.6%
ScoreGPT-5.6 Luna (max)OpenAIIntelligence index43.4→37.5-14%
ScoreGPT-5.6 Luna (max)OpenAIAgentic index42.9→42.7-0.5%
ScoreGPT-5.6 Luna (medium)OpenAIIntelligence index30.2→25.8-15%
ScoreGPT-5.6 Luna (medium)OpenAIAgentic index25.5→25.4-0.4%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIIntelligence index19.3→16.8-13%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIAgentic index15.3→15.2-0.7%
ScoreGPT-5.6 Luna (xhigh)OpenAIIntelligence index41.6→34.8-16%
ScoreGPT-5.6 Luna (xhigh)OpenAIAgentic index39.7→39.5-0.5%
ScoreGPT-5.6 Sol (high)OpenAIIntelligence index48.3→42.5-12%
ScoreGPT-5.6 Sol (high)OpenAIAgentic index44.9→44.7-0.4%
ScoreGPT-5.6 Sol (low)OpenAIIntelligence index40.8→33.8-17%
ScoreGPT-5.6 Sol (low)OpenAIAgentic index32.9→32.6-0.9%
ScoreGPT-5.6 Sol (max)OpenAIIntelligence index51.3→47.1-8.2%
ScoreGPT-5.6 Sol (max)OpenAIAgentic index50.7→50.5-0.4%
ScoreGPT-5.6 Sol (medium)OpenAIIntelligence index46→39.5-14%
ScoreGPT-5.6 Sol (medium)OpenAIAgentic index40.8→40.5-0.7%
ScoreGPT-5.6 Sol (Non-reasoning)OpenAIIntelligence index32.9→28.3-14%
ScoreGPT-5.6 Sol (Non-reasoning)OpenAIAgentic index29.5→29.4-0.3%
ScoreGPT-5.6 Sol (xhigh)OpenAIIntelligence index49.8→44.1-11%
ScoreGPT-5.6 Sol (xhigh)OpenAIAgentic index48.1→47.8-0.6%
ScoreGPT-5.6 Terra (high)OpenAIIntelligence index41.3→34.5-16%
ScoreGPT-5.6 Terra (high)OpenAIAgentic index37.8→37.6-0.5%
ScoreGPT-5.6 Terra (low)OpenAIIntelligence index32.4→27.9-14%
ScoreGPT-5.6 Terra (max)OpenAIIntelligence index46.8→42.3-9.6%
ScoreGPT-5.6 Terra (max)OpenAIAgentic index43.9→43.7-0.5%
ScoreGPT-5.6 Terra (medium)OpenAIIntelligence index37.2→32.8-12%
ScoreGPT-5.6 Terra (medium)OpenAIAgentic index31.6→31.4-0.6%
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIIntelligence index26.3→22.3-15%
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIAgentic index25.1→24.9-0.8%
ScoreGPT-5.6 Terra (xhigh)OpenAIIntelligence index44.4→38.2-14%
ScoreGPT-5.6 Terra (xhigh)OpenAIAgentic index42.4→42.2-0.5%
ScoreGPT-6 Astra (high)OpenAIIntelligence index53.4→51-4.5%
ScoreGPT-6 Astra (high)OpenAIAgentic index48.9→48.7-0.4%
ScoreGPT-6 Astra (low)OpenAIIntelligence index49.3→46-6.7%
ScoreGPT-6 Astra (low)OpenAIAgentic index39.7→39.5-0.5%
ScoreGPT-6 Astra (max)OpenAIIntelligence index54.7→52.8-3.5%
ScoreGPT-6 Astra (max)OpenAIAgentic index51.6→51.5-0.2%
ScoreGPT-6 Astra (medium)OpenAIIntelligence index52.2→49.7-4.8%
ScoreGPT-6 Astra (medium)OpenAIAgentic index46.5→46.3-0.4%
ScoreGPT-6 Astra (Non-reasoning)OpenAIIntelligence index47.8→45.2-5.4%
ScoreGPT-6 Astra (Non-reasoning)OpenAIAgentic index48→47.8-0.4%
ScoreGPT-6 Astra (xhigh)OpenAIIntelligence index54.3→52.5-3.3%
ScoreGPT-6 Astra (xhigh)OpenAIAgentic index50.8→50.6-0.4%
Scoregpt-oss-120b (high)OpenAIIntelligence index15.6→12.3-21%
Scoregpt-oss-120b (high)OpenAIAgentic index6.3→6.2-1.6%
Scoregpt-oss-120b (low)OpenAIIntelligence index8.9→10.2+15%
Scoregpt-oss-20b (high)OpenAIIntelligence index10.8→9-17%
Scoregpt-oss-20b (low)OpenAIIntelligence index8.4→10+19%
ScoreGranite 3.3 8B (Non-reasoning)IBMIntelligence index1→4.9+390%
ScoreGranite 4.0 1BIBMIntelligence index1→5+400%
ScoreGranite 4.0 350MIBMIntelligence index1→4.8+380%
ScoreGranite 4.0 H 1BIBMIntelligence index1→5.2+420%
ScoreGranite 4.0 H 350MIBMIntelligence index1→4.8+380%
ScoreGranite 4.0 H SmallIBMIntelligence index1→6+500%
ScoreGranite 4.0 MicroIBMIntelligence index1→5.1+410%
ScoreGranite 4.1 30BIBMIntelligence index3.2→7.4+131%
ScoreGranite 4.1 3BIBMIntelligence index1→5.9+490%
ScoreGranite 4.1 8BIBMIntelligence index1.1→6.6+500%
ScoreGranite 4.2 30BIBMIntelligence index16.5→14.8-10%
ScoreGranite 4.2 3BIBMIntelligence index10.7→9.1-15%
ScoreGranite 4.2 8BIBMIntelligence index13.9→12.4-11%
ScoreGranite 4.2 8BIBMAgentic index3.8→3.7-2.6%
ScoreGrok 2 (Dec '24)SpaceXAIIntelligence index2.4→7.1+196%
ScoreGrok 3SpaceXAIIntelligence index12.3→12.1-1.6%
ScoreGrok 3 mini Reasoning (high)SpaceXAIIntelligence index16.2→14.6-9.9%
ScoreGrok 3 Reasoning BetaSpaceXAIIntelligence index9.2→10.4+13%
ScoreGrok 4SpaceXAIIntelligence index26.5→22.5-15%
ScoreGrok 4 Fast (Non-reasoning)SpaceXAIIntelligence index10.5→11.1+5.7%
ScoreGrok 4 Fast (Reasoning)SpaceXAIIntelligence index20.8→17.9-14%
ScoreGrok 4.1 Fast (Non-reasoning)SpaceXAIIntelligence index10.9→11.3+3.7%
ScoreGrok 4.1 Fast (Reasoning)SpaceXAIIntelligence index23.9→20.4-15%
ScoreGrok 4.20 0309 (Non-reasoning)SpaceXAIIntelligence index16.2→14.6-9.9%
ScoreGrok 4.20 0309 (Reasoning)SpaceXAIIntelligence index29.5→25.2-15%
ScoreGrok 4.20 0309 v2 (Non-reasoning)SpaceXAIIntelligence index15.6→14.2-9.0%
ScoreGrok 4.20 0309 v2 (Reasoning)SpaceXAIIntelligence index30→25.7-14%
ScoreGrok 4.3 (high)SpaceXAIIntelligence index29.9→25.4-15%
ScoreGrok 4.3 (high)SpaceXAIAgentic index17.3→17.2-0.6%
ScoreGrok 4.3 (low)SpaceXAIIntelligence index28.5→24.3-15%
ScoreGrok 4.3 (medium)SpaceXAIIntelligence index29.1→24.8-15%
ScoreGrok 4.3 (Non-reasoning)SpaceXAIIntelligence index17.5→14.5-17%
ScoreGrok 4.3 (Non-reasoning)SpaceXAIAgentic indexFirst measured→15.8
ScoreGrok 4.5 (high)SpaceXAIIntelligence index45.5→39.1-14%
ScoreGrok 4.5 (high)SpaceXAIAgentic index42.4→42.1-0.7%
ScoreGrok 4.6 (high)SpaceXAIIntelligence index50.6→44.4-12%
ScoreGrok 4.6 (high)SpaceXAIAgentic index53.6→53.4-0.4%
ScoreGrok 4.6 (low)SpaceXAIIntelligence index41.6→35.4-15%
ScoreGrok 4.6 (low)SpaceXAIAgentic index42.4→42.2-0.5%
ScoreGrok 4.6 (medium)SpaceXAIIntelligence index48.8→43-12%
ScoreGrok 4.6 (medium)SpaceXAIAgentic index51.4→51.2-0.4%
ScoreGrok 4.6 (xhigh)SpaceXAIIntelligence index49.9→44.3-11%
ScoreGrok 4.6 (xhigh)SpaceXAIAgentic index53→52.7-0.6%
ScoreGrok BetaSpaceXAIIntelligence index1.9→6.9+263%
ScoreGrok Build 0.1 0616SpaceXAIIntelligence index31.7→27.2-14%
ScoreGrok Code Fast 1SpaceXAIIntelligence index15.4→14.1-8.4%
ScoreGrok-1SpaceXAIIntelligence index1→6.3+530%
ScoreHermes 3 - Llama-3.1 70BNous ResearchIntelligence index1→6+500%
ScoreHermes 4 - Llama-3.1 405B (Non-reasoning)Nous ResearchIntelligence index3.2→7.4+131%
ScoreHermes 4 - Llama-3.1 405B (Reasoning)Nous ResearchIntelligence index3.4→7.5+121%
ScoreHermes 4 - Llama-3.1 70B (Non-reasoning)Nous ResearchIntelligence index1.3→6.7+415%
ScoreHermes 4 - Llama-3.1 70B (Reasoning)Nous ResearchIntelligence index4.3→7.9+84%
ScoreHy3TencentIntelligence index32.4→25.8-20%
ScoreHy3TencentAgentic index25.7→25.6-0.4%
ScoreHy3-preview (Non-reasoning)TencentIntelligence index19.6→17-13%
ScoreHy3-preview (Reasoning)TencentIntelligence index26.8→22.7-15%
ScoreHyperCLOVA X SEED Think (32B)NaverIntelligence index11→11.4+3.6%
ScoreInkling (xhigh)Thinking MachinesIntelligence index32.2→25.5-21%
ScoreInkling (xhigh)Thinking MachinesAgentic index24.4→24.3-0.4%
ScoreInkling SmallThinking MachinesIntelligence index32.2→26.1-19%
ScoreInkling SmallThinking MachinesAgentic index25.1→25-0.4%
ScoreINTELLECT-3 (based on GLM-4.5-Air)Prime IntellectIntelligence index9.6→10.6+10%
ScoreJamba 1.5 LargeAI21 LabsIntelligence index1→6+500%
ScoreJamba 1.5 MiniAI21 LabsIntelligence index1→5.2+420%
ScoreJamba 1.6 LargeAI21 LabsIntelligence index1→6+500%
ScoreJamba 1.6 MiniAI21 LabsIntelligence index1→5.2+420%
ScoreJamba 1.7 LargeAI21 LabsIntelligence index1→6.1+510%
ScoreJamba 1.7 MiniAI21 LabsIntelligence index1→5.2+420%
ScoreJamba Reasoning 3BAI21 LabsIntelligence index1→5.7+470%
ScoreJT-35B-FlashChina MobileIntelligence index21.8→18.7-14%
ScoreJT-4.1 Flash 236B A21BChina MobileIntelligence index31.8→27.3-14%
ScoreJT-MINIChina MobileIntelligence index12.4→12.2-1.6%
ScoreK-EXAONE (Non-reasoning)LG AI ResearchIntelligence index10.7→11.2+4.7%
ScoreK-EXAONE (Reasoning)LG AI ResearchIntelligence index15.8→14.4-8.9%
ScoreK-EXAONE 2.0 0803LG AI ResearchIntelligence index23.1→19.7-15%
ScoreK2 Horizon 375B A23BMBZUAI Institute of Foundation ModelsIntelligence index37.8→33.9-10%
ScoreK2 Horizon 375B A23BMBZUAI Institute of Foundation ModelsAgentic index41.2→40.9-0.7%
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsIntelligence index11.2→11.5+2.7%
ScoreK2-V2 (high)MBZUAI Institute of Foundation ModelsIntelligence index8.3→9.9+19%
ScoreK2-V2 (low)MBZUAI Institute of Foundation ModelsIntelligence index2.9→7.3+152%
ScoreK2-V2 (medium)MBZUAI Institute of Foundation ModelsIntelligence index6.6→9+36%
ScoreKAT Coder Pro V2KwaiKATIntelligence index25.6→21.7-15%
ScoreKAT-Coder-Pro V1KwaiKATIntelligence index21.7→18.6-14%
ScoreKimi K2KimiIntelligence index13.3→12.7-4.5%
ScoreKimi K2 0905KimiIntelligence index17.2→15.3-11%
ScoreKimi K2 ThinkingKimiIntelligence index25.9→22-15%
ScoreKimi K2.5 (Non-reasoning)KimiIntelligence index22.8→19.4-15%
ScoreKimi K2.5 (Reasoning)KimiIntelligence index27.6→23.5-15%
ScoreKimi K2.6KimiIntelligence index35.8→31.3-13%
ScoreKimi K2.6KimiAgentic index22.2→22.1-0.5%
ScoreKimi K2.6 (Non-reasoning)KimiIntelligence index27.7→23.6-15%
ScoreKimi K2.7 CodeKimiIntelligence index32.7→26.3-20%
ScoreKimi K2.7 CodeKimiAgentic index22.6→22.5-0.4%
ScoreKimi K3 (low)KimiIntelligence index37.1→34.5-7.0%
ScoreKimi K3 (low)KimiAgentic index30.8→30.7-0.3%
ScoreKimi K3 (max)KimiIntelligence index50.2→43.8-13%
ScoreKimi K3 (max)KimiAgentic index50.9→50.6-0.6%
ScoreKimi Linear 48B A3B InstructKimiIntelligence index2.9→7.3+152%
ScoreLFM 40BLiquid AIIntelligence index1→5.4+440%
ScoreLFM2 1.2BLiquid AIIntelligence index1→4.8+380%
ScoreLFM2 2.6BLiquid AIIntelligence index1→5.2+420%
ScoreLFM2 24B A2BLiquid AIIntelligence index1→5.9+490%
ScoreLFM2 8B A1BLiquid AIIntelligence index1→4.9+390%
ScoreLFM2.5-1.2B-InstructLiquid AIIntelligence index1→5.2+420%
ScoreLFM2.5-1.2B-ThinkingLiquid AIIntelligence index1→5.2+420%
ScoreLFM2.5-2.6BLiquid AIIntelligence index5.3→8.4+58%
ScoreLFM2.5-8B-A1BLiquid AIIntelligence index2.7→7.2+167%
ScoreLFM2.5-VL-1.6BLiquid AIIntelligence index1→4.8+380%
ScoreLing 2.6 FlashInclusionAIIntelligence index7.9→9.7+23%
ScoreLing 3.0 FlashInclusionAIIntelligence index27.4→24.9-9.1%
ScoreLing 3.0 FlashInclusionAIAgentic index21.1→21-0.5%
ScoreLing 3.0 TinyInclusionAIIntelligence index15.7→11.9-24%
ScoreLing-1TInclusionAIIntelligence index6.9→9.2+33%
ScoreLing-2.6-1TInclusionAIIntelligence index19.6→17-13%
ScoreLing-flash-2.0InclusionAIIntelligence index4.1→7.8+90%
ScoreLing-mini-2.0InclusionAIIntelligence index1→5.5+450%
ScoreLlama 2 Chat 13BMetaIntelligence index1→5.3+430%
ScoreLlama 2 Chat 70BMetaIntelligence index1→5.3+430%
ScoreLlama 2 Chat 7BMetaIntelligence index1→5.7+470%
ScoreLlama 3 Instruct 70BMetaIntelligence index1→5.5+450%
ScoreLlama 3 Instruct 8BMetaIntelligence index1→4.8+380%
ScoreLlama 3.1 Instruct 405BMetaIntelligence index2.9→7.3+152%
ScoreLlama 3.1 Instruct 70BMetaIntelligence index1.2→6.6+450%
ScoreLlama 3.1 Instruct 8BMetaIntelligence index2→6.9+245%
ScoreLlama 3.1 Nemotron Instruct 70BNVIDIAIntelligence index2.1→6.9+229%
ScoreLlama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIAIntelligence index2.9→7.3+152%
ScoreLlama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIAIntelligence index3.4→7.5+121%
ScoreLlama 3.1 Tulu3 405BAllen Institute for AIIntelligence index2.7→7.2+167%
ScoreLlama 3.2 Instruct 11B (Vision)MetaIntelligence index1→5.4+440%
ScoreLlama 3.2 Instruct 1BMetaIntelligence index1→4.8+380%
ScoreLlama 3.2 Instruct 3BMetaIntelligence index1→5.7+470%
ScoreLlama 3.2 Instruct 90B (Vision)MetaIntelligence index1→6.4+540%
ScoreLlama 3.3 Instruct 70BMetaIntelligence index3.7→7.7+108%
ScoreLlama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIAIntelligence index2.8→7.3+161%
ScoreLlama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIAIntelligence index6.4→8.9+39%
ScoreLlama 4 MaverickMetaIntelligence index10.3→9.3-9.7%
ScoreLlama 4 ScoutMetaIntelligence index4.6→6.5+41%
ScoreLlama 65BMetaIntelligence index1→5+400%
ScoreLlama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIAIntelligence index3→7.4+147%
ScoreLlama Nemotron Super 49B v1.5 (Reasoning)NVIDIAIntelligence index6.6→9+36%
ScoreLongCat 2.0LongCatIntelligence index25.8→19.7-24%
ScoreLongCat 2.0LongCatAgentic indexFirst measured→15.9
ScoreLongCat Flash LiteLongCatIntelligence index11.2→11.5+2.7%
ScoreMagistral Medium 1MistralIntelligence index6.7→9.1+36%
ScoreMagistral Small 1MistralIntelligence index5→8.2+64%
ScoreMagistral Small 1.2MistralIntelligence index5.7→8.6+51%
ScoreMercury 2InceptionIntelligence index14.9→11.5-23%
ScoreMercury 2InceptionAgentic index4.1→4-2.4%
ScoreMi:dm K 2.5 ProKorea TelecomIntelligence index10.4→11+5.8%
ScoreMiMo-V2-Flash (Feb 2026)XiaomiIntelligence index26.4→22.4-15%
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiIntelligence index18.3→16-13%
ScoreMiMo-V2-Flash (Reasoning)XiaomiIntelligence index24.5→20.8-15%
ScoreMiMo-V2-OmniXiaomiIntelligence index28.1→23.9-15%
ScoreMiMo-V2-Omni-0327XiaomiIntelligence index29.4→25.1-15%
ScoreMiMo-V2-ProXiaomiIntelligence index33.1→28.6-14%
ScoreMiMo-V2.5XiaomiIntelligence index28.2→22.3-21%
ScoreMiMo-V2.5-ProXiaomiIntelligence index32.6→26.4-19%
ScoreMiMo-V2.5-Pro (Non-reasoning)XiaomiIntelligence index21.3→18.3-14%
ScoreMiniCPM-V 4.6 1.3BOpenBMBIntelligence index1→5.7+470%
ScoreMiniCPM5-1B (Non-reasoning)OpenBMBIntelligence index6→8.7+45%
ScoreMiniCPM5-1B (Reasoning)OpenBMBIntelligence index6.2→8.8+42%
ScoreMiniCPM5-2BOpenBMBIntelligence index15→14.3-4.7%
ScoreMiniMax M1 40kMiniMaxIntelligence index8.5→10+18%
ScoreMiniMax M1 80kMiniMaxIntelligence index11.6→11.7+0.9%
ScoreMiniMax-M2MiniMaxIntelligence index21.7→18.6-14%
ScoreMiniMax-M2.1MiniMaxIntelligence index24.6→20.9-15%
ScoreMiniMax-M2.5MiniMaxIntelligence index26.8→22.8-15%
ScoreMiniMax-M2.7MiniMaxIntelligence index30.1→23.2-23%
ScoreMiniMax-M2.7MiniMaxAgentic index16.9→16.8-0.6%
ScoreMiniMax-M3MiniMaxIntelligence index35.7→29.6-17%
ScoreMiniMax-M3MiniMaxAgentic index31→30.8-0.6%
ScoreMinistral 3 14BMistralIntelligence index5.5→6+9.1%
ScoreMinistral 3 3BMistralIntelligence index1.7→4.8+182%
ScoreMinistral 3 8BMistralIntelligence index3.5→5.5+57%
ScoreMistral 7B InstructMistralIntelligence index1→5+400%
ScoreMistral Large (Feb '24)MistralIntelligence index1→5.8+480%
ScoreMistral Large 2 (Jul '24)MistralIntelligence index1.7→6.8+300%
ScoreMistral Large 2 (Nov '24)MistralIntelligence index3.5→7.6+117%
ScoreMistral Large 3MistralIntelligence index11.1→9.7-13%
ScoreMistral Large 3MistralAgentic index2.5→2.4-4.0%
ScoreMistral MediumMistralIntelligence index1→5.5+450%
ScoreMistral Medium 3MistralIntelligence index6.7→9+34%
ScoreMistral Medium 3.1MistralIntelligence index8.4→9.9+18%
ScoreMistral Medium 3.1MistralAgentic index3.3→3.1-6.1%
ScoreMistral Medium 3.5MistralIntelligence index22.7→14.9-34%
ScoreMistral SabaMistralIntelligence index1→6.5+550%
ScoreMistral Small (Feb '24)MistralIntelligence index1→5.5+450%
ScoreMistral Small (Sep '24)MistralIntelligence index1→5.8+480%
ScoreMistral Small 3MistralIntelligence index1.4→6.7+379%
ScoreMistral Small 3.1MistralIntelligence index10.3→7.4-28%
ScoreMistral Small 3.2MistralIntelligence index5→7+40%
ScoreMistral Small 4 (Non-reasoning)MistralIntelligence index6.6→9+36%
ScoreMistral Small 4 (Reasoning)MistralIntelligence index12.8→11.5-10%
ScoreMistral Small 4 (Reasoning)MistralAgentic index1.5→1.4-6.7%
ScoreMixtral 8x22B InstructMistralIntelligence index1→5.7+470%
ScoreMixtral 8x7B InstructMistralIntelligence index1→5.1+410%
ScoreMolmo 7B-DAllen Institute for AIIntelligence index1→5.6+460%
ScoreMolmo2-8BAllen Institute for AIIntelligence index1→5+400%
ScoreMotif 3Motif TechnologiesIntelligence index38→33.6-12%
ScoreMotif 3 (Beta)Motif TechnologiesIntelligence index36.8→32.3-12%
ScoreMotif-2-12.7B-ReasoningMotif TechnologiesIntelligence index7→9.2+31%
ScoreMuse Glimmer (high)MetaIntelligence index24.4→18.1-26%
ScoreMuse SparkMetaIntelligence index35.8→31.3-13%
ScoreMuse Spark 1.1 (xhigh)MetaIntelligence index41.2→34.3-17%
ScoreMuse Spark 1.1 (xhigh)MetaAgentic index27.8→27.5-1.1%
ScoreMuse Spark 1.2 (xhigh)MetaIntelligence index46.8→39.8-15%
ScoreMuse Spark 1.2 (xhigh)MetaAgentic index44.2→44-0.5%
ScoreMuse Spark 1.3 (xhigh)MetaIntelligence index51.6→45.2-12%
ScoreMuse Spark 1.3 (xhigh)MetaAgentic index52.1→51.8-0.6%
ScoreNanbeige4.1-3BNanbeigeIntelligence index5.3→8.4+58%
ScoreNemotron 3 Nano Omni 30B A3B ReasoningNVIDIAIntelligence index9→10.3+14%
ScoreNemotron 3 Super 120B A12B (Reasoning)NVIDIAIntelligence index18.6→13.6-27%
ScoreNemotron 3 Super 120B A12B (Reasoning)NVIDIAAgentic index4.2→4.1-2.4%
ScoreNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAIntelligence index29.3→23.4-20%
ScoreNemotron 3.5 LightningNVIDIAIntelligence index16.4→13.6-17%
ScoreNemotron 3.5 LightningNVIDIAAgentic indexFirst measured→6.1
ScoreNemotron Cascade 2 30B A3BNVIDIAIntelligence index11.5→11.7+1.7%
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIIntelligence index32.7→28.2-14%
ScoreNorth Mini CodeCohereIntelligence index13.5→12.8-5.2%
ScoreNova 2.0 Lite (high)AmazonIntelligence index14.3→13.4-6.3%
ScoreNova 2.0 Lite (low)AmazonIntelligence index11.7→11.8+0.9%
ScoreNova 2.0 Lite (medium)AmazonIntelligence index12.9→12.5-3.1%
ScoreNova 2.0 Lite (Non-reasoning)AmazonIntelligence index6→8.7+45%
ScoreNova 2.0 Omni (low)AmazonIntelligence index10.6→11.1+4.7%
ScoreNova 2.0 Omni (medium)AmazonIntelligence index14.7→13.6-7.5%
ScoreNova 2.0 Omni (Non-reasoning)AmazonIntelligence index4.8→8.2+71%
ScoreNova 2.0 Pro Preview (low)AmazonIntelligence index13.4→12.8-4.5%
ScoreNova 2.0 Pro Preview (medium)AmazonIntelligence index15.5→14.2-8.4%
ScoreNova 2.0 Pro Preview (Non-reasoning)AmazonIntelligence index8.5→10+18%
ScoreNova LiteAmazonIntelligence index1.4→6.7+379%
ScoreNova MicroAmazonIntelligence index1→5.9+490%
ScoreNova PremierAmazonIntelligence index6.9→9.2+33%
ScoreNova ProAmazonIntelligence index2.1→7+233%
ScoreNVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIAIntelligence index1.8→6.8+278%
ScoreNVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIAIntelligence index8.6→8.9+3.5%
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAIntelligence index3.1→7.4+139%
ScoreNVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIAIntelligence index1→5.8+480%
ScoreNVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIAIntelligence index3.3→7.5+127%
ScoreNVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIAIntelligence index1.8→6.8+278%
ScoreNVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIAIntelligence index3.2→7.4+131%
Scoreo1OpenAIIntelligence index17.1→15.2-11%
Scoreo1-miniOpenAIIntelligence index8.1→9.8+21%
Scoreo1-previewOpenAIIntelligence index11→11.4+3.6%
Scoreo1-proOpenAIIntelligence index12.8→12.4-3.1%
Scoreo3OpenAIIntelligence index23.7→20.2-15%
Scoreo3-miniOpenAIIntelligence index12.9→12.5-3.1%
Scoreo3-mini (high)OpenAIIntelligence index9.6→11+15%
Scoreo3-proOpenAIIntelligence index25.7→21.9-15%
Scoreo4-mini (high)OpenAIIntelligence index19.1→16.7-13%
ScoreOLMo 2 32BAllen Institute for AIIntelligence index1→6+500%
ScoreOLMo 2 7BAllen Institute for AIIntelligence index1→5.6+460%
ScoreOlmo 3 32B ThinkAllen Institute for AIIntelligence index1→6.5+550%
ScoreOlmo 3 7B InstructAllen Institute for AIIntelligence index1→5.2+420%
ScoreOlmo 3 7B ThinkAllen Institute for AIIntelligence index1→5.6+460%
ScoreOlmo 3.1 32B InstructAllen Institute for AIIntelligence index1→6.5+550%
ScoreOlmo 3.1 32B ThinkAllen Institute for AIIntelligence index2.5→7.1+184%
ScoreOpenChat 3.5 (1210)OpenChatIntelligence index1→5.3+430%
ScorePALM-2GoogleIntelligence index1→5.4+440%
ScorePhi-3 Mini Instruct 3.8BMicrosoftIntelligence index1→5.8+480%
ScorePhi-4MicrosoftIntelligence index1→5.9+490%
ScorePhi-4 Mini InstructMicrosoftIntelligence index1→6.3+530%
ScorePhi-4 Multimodal InstructMicrosoftIntelligence index1→5.8+480%
ScorePixtral LargeMistralIntelligence index2.5→7.1+184%
ScoreQuasar 438B (max, based on GLM-5.2)Multiverse ComputingIntelligence index34.2→27.1-21%
ScoreQuasar 438B (max, based on GLM-5.2)Multiverse ComputingAgentic index32.8→32.7-0.3%
ScoreQwen Chat 14BAlibabaIntelligence index1→5+400%
ScoreQwen Chat 72BAlibabaIntelligence index1→5.4+440%
ScoreQwen1.5 Chat 110BAlibabaIntelligence index1→5.7+470%
ScoreQwen2 Instruct 72BAlibabaIntelligence index1→6.3+530%
ScoreQwen2.5 Coder Instruct 32BAlibabaIntelligence index1.6→6.7+319%
ScoreQwen2.5 Coder Instruct 7BAlibabaIntelligence index1→5.8+480%
ScoreQwen2.5 Instruct 32BAlibabaIntelligence index1.9→6.9+263%
ScoreQwen2.5 Instruct 72BAlibabaIntelligence index3.9→7.7+97%
ScoreQwen2.5 MaxAlibabaIntelligence index4.5→8+78%
ScoreQwen2.5 TurboAlibabaIntelligence index1→6.4+540%
ScoreQwen3 0.6B (Non-reasoning)AlibabaIntelligence index1→4.8+380%
ScoreQwen3 0.6B (Reasoning)AlibabaIntelligence index1→4.8+380%
ScoreQwen3 1.7B (Non-reasoning)AlibabaIntelligence index1→4.9+390%
ScoreQwen3 1.7B (Reasoning)AlibabaIntelligence index1→5.2+420%
ScoreQwen3 14B (Non-reasoning)AlibabaIntelligence index1.5→6.7+347%
ScoreQwen3 14B (Reasoning)AlibabaIntelligence index7.1→6.4-9.9%
ScoreQwen3 235B A22B (Non-reasoning)AlibabaIntelligence index5.1→8.3+63%
ScoreQwen3 235B A22B (Reasoning)AlibabaIntelligence index7.6→9.5+25%
ScoreQwen3 235B A22B 2507 (Reasoning)AlibabaIntelligence index13.1→12.7-3.1%
ScoreQwen3 235B A22B 2507 InstructAlibabaIntelligence index12.1→12-0.8%
ScoreQwen3 30B A3B (Non-reasoning)AlibabaIntelligence index1.3→6.6+408%
ScoreQwen3 30B A3B (Reasoning)AlibabaIntelligence index3.6→7.6+111%
ScoreQwen3 30B A3B 2507 (Reasoning)AlibabaIntelligence index8.6→9.8+14%
ScoreQwen3 30B A3B 2507 InstructAlibabaIntelligence index3.4→7.5+121%
ScoreQwen3 32B (Non-reasoning)AlibabaIntelligence index3→7.3+143%
ScoreQwen3 32B (Reasoning)AlibabaIntelligence index8→7.2-10.0%
ScoreQwen3 4B (Non-reasoning)AlibabaIntelligence index1.2→6.6+450%
ScoreQwen3 4B (Reasoning)AlibabaIntelligence index2.7→7.2+167%
ScoreQwen3 4B 2507 (Reasoning)AlibabaIntelligence index6.2→8.8+42%
ScoreQwen3 4B 2507 InstructAlibabaIntelligence index1.6→6.7+319%
ScoreQwen3 8B (Non-reasoning)AlibabaIntelligence index1→6+500%
ScoreQwen3 8B (Reasoning)AlibabaIntelligence index2.8→5.2+86%
ScoreQwen3 Coder 30B A3B InstructAlibabaIntelligence index7.7→9.6+25%
ScoreQwen3 Coder NextAlibabaIntelligence index14.1→10.1-28%
ScoreQwen3 Coder NextAlibabaAgentic index3.7→3.6-2.7%
ScoreQwen3 MaxAlibabaIntelligence index17.6→15.6-11%
ScoreQwen3 Max (Preview)AlibabaIntelligence index13→12.6-3.1%
ScoreQwen3 Max ThinkingAlibabaIntelligence index25→21.3-15%
ScoreQwen3 Max Thinking (Preview)AlibabaIntelligence index18.6→16.3-12%
ScoreQwen3 Next 80B A3B (Reasoning)AlibabaIntelligence index10.7→11.2+4.7%
ScoreQwen3 Next 80B A3B InstructAlibabaIntelligence index7.8→9.6+23%
ScoreQwen3 Omni 30B A3B (Reasoning)AlibabaIntelligence index4→7.8+95%
ScoreQwen3 Omni 30B A3B InstructAlibabaIntelligence index1→6+500%
ScoreQwen3 VL 235B A22B (Reasoning)AlibabaIntelligence index14.4→13.4-6.9%
ScoreQwen3 VL 235B A22B InstructAlibabaIntelligence index8.4→9.9+18%
ScoreQwen3 VL 30B A3B (Reasoning)AlibabaIntelligence index7.5→9.5+27%
ScoreQwen3 VL 30B A3B InstructAlibabaIntelligence index4.3→7.9+84%
ScoreQwen3 VL 32B InstructAlibabaIntelligence index5.3→8.4+58%
ScoreQwen3 VL 4B (Reasoning)AlibabaIntelligence index2.3→7+204%
ScoreQwen3 VL 4B InstructAlibabaIntelligence index1→5.7+470%
ScoreQwen3 VL 8B (Reasoning)AlibabaIntelligence index4.9→8.2+67%
ScoreQwen3 VL 8B InstructAlibabaIntelligence index2.8→7.3+161%
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaIntelligence index1→5.4+440%
ScoreQwen3.5 0.8B (Reasoning)AlibabaIntelligence index1→6.1+510%
ScoreQwen3.5 122B A10B (Non-reasoning)AlibabaIntelligence index20.6→17.7-14%
ScoreQwen3.5 122B A10B (Reasoning)AlibabaIntelligence index24.8→16.2-35%
ScoreQwen3.5 122B A10B (Reasoning)AlibabaAgentic index9.8→9.6-2.0%
ScoreQwen3.5 27B (Non-reasoning)AlibabaIntelligence index22.7→19.4-15%
ScoreQwen3.5 27B (Reasoning)AlibabaIntelligence index26.9→22.9-15%
ScoreQwen3.5 2B (Non-reasoning)AlibabaIntelligence index1→6.2+520%
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index2.1→6.9+229%
ScoreQwen3.5 35B A3B (Non-reasoning)AlibabaIntelligence index17→15.1-11%
ScoreQwen3.5 35B A3B (Reasoning)AlibabaIntelligence index22.6→19.3-15%
ScoreQwen3.5 397B A17B (Non-reasoning)AlibabaIntelligence index25.2→21.4-15%
ScoreQwen3.5 397B A17B (Reasoning)AlibabaIntelligence index26.1→19.1-27%
ScoreQwen3.5 4B (Non-reasoning)AlibabaIntelligence index10→10.8+8.0%
ScoreQwen3.5 4B (Reasoning)AlibabaIntelligence index13.9→13.1-5.8%
ScoreQwen3.5 9B (Non-reasoning)AlibabaIntelligence index14.1→13.3-5.7%
ScoreQwen3.5 9B (Reasoning)AlibabaIntelligence index14.8→13.7-7.4%
ScoreQwen3.5 Omni FlashAlibabaIntelligence index12.9→12.5-3.1%
ScoreQwen3.5 Omni PlusAlibabaIntelligence index24→20.4-15%
ScoreQwen3.6 27B (Non-reasoning)AlibabaIntelligence index23.3→19.8-15%
ScoreQwen3.6 27B (Reasoning)AlibabaIntelligence index29→21.9-24%
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaIntelligence index17.1→15.2-11%
ScoreQwen3.6 35B A3B (Reasoning)AlibabaIntelligence index26.2→22.3-15%
ScoreQwen3.6 35B A3B (Reasoning)AlibabaAgentic index15.1→15-0.7%
ScoreQwen3.6 Max PreviewAlibabaIntelligence index32.9→28.4-14%
ScoreQwen3.6 PlusAlibabaIntelligence index31.5→27-14%
ScoreQwen3.7 MaxAlibabaIntelligence index36.6→29.9-18%
ScoreQwen3.7 PlusAlibabaIntelligence index31.9→25.8-19%
ScoreQwen3.7 PlusAlibabaAgentic index19.8→19.7-0.5%
ScoreQwen3.8 2.4T A95BAlibabaIntelligence index46.7→40-14%
ScoreQwen3.8 2.4T A95BAlibabaAgentic index50.7→50.4-0.6%
ScoreQwen3.8 27B (low)AlibabaIntelligence index33.8→29.2-14%
ScoreQwen3.8 27B (low)AlibabaAgentic indexFirst measured→39.9
ScoreQwen3.8 27B (medium)AlibabaIntelligence index35.2→30.7-13%
ScoreQwen3.8 27B (Non-reasoning)AlibabaIntelligence index26.5→22.4-15%
ScoreQwen3.8 27B (xhigh)AlibabaIntelligence index41.4→33.9-18%
ScoreQwen3.8 27B (xhigh)AlibabaAgentic index46.8→46.5-0.6%
ScoreQwen3.8 MaxAlibabaIntelligence index46.9→40.3-14%
ScoreQwen3.8 MaxAlibabaAgentic index49.9→49.6-0.6%
ScoreQwen3.8-Flash-NextAlibabaIntelligence index45.6→42.2-7.5%
ScoreQwQ 32BAlibabaIntelligence index7.5→9.5+27%
ScoreQwQ 32B-PreviewAlibabaIntelligence index3.6→7.6+111%
ScoreR1 1776PerplexityIntelligence index1→6.4+540%
ScoreReka Flash (Sep '24)Reka AIIntelligence index1→6.4+540%
ScoreReka Flash 3Reka AIIntelligence index1→5.6+460%
ScoreRing-1TInclusionAIIntelligence index10.2→10.9+6.9%
ScoreRing-2.6-1TInclusionAIIntelligence index23.7→17.3-27%
ScoreRing-2.6-1TInclusionAIAgentic index13→12.9-0.8%
ScoreRing-flash-2.0InclusionAIIntelligence index2.6→7.2+177%
ScoreSarvam 105B (high)SarvamIntelligence index6.2→8.8+42%
ScoreSarvam 30B (high)SarvamIntelligence index1.1→6.6+500%
ScoreSarvam M (Reasoning, based on Mistral Small 3.1)SarvamIntelligence index1→5.3+430%
ScoreSeed-OSS-36B-InstructByteDance SeedIntelligence index12.2→12.1-0.8%
ScoreSolar MiniUpstageIntelligence index1→6.4+540%
ScoreSolar Open 100B (Reasoning)UpstageIntelligence index9.2→10.4+13%
ScoreSolar Open2 250BUpstageIntelligence index29→24.7-15%
ScoreSolar Pro 2 (Non-reasoning)UpstageIntelligence index2.2→7+218%
ScoreSolar Pro 2 (Preview) (Non-reasoning)UpstageIntelligence index4.3→7.9+84%
ScoreSolar Pro 2 (Preview) (Reasoning)UpstageIntelligence index6.7→9.1+36%
ScoreSolar Pro 2 (Reasoning)UpstageIntelligence index3.3→7.5+127%
ScoreSolar Pro 3UpstageIntelligence index9.4→7.8-17%
ScoreSolar Pro 4UpstageIntelligence index32.7→28.2-14%
ScoreSonarPerplexityIntelligence index3.8→7.7+103%
ScoreSonar ProPerplexityIntelligence index3.6→7.6+111%
ScoreSonar ReasoningPerplexityIntelligence index5.9→8.7+47%
ScoreStep 3.5 FlashStepFunIntelligence index19.1→16.6-13%
ScoreStep 3.5 Flash 2603StepFunIntelligence index19.5→17-13%
ScoreStep 3.7 FlashStepFunIntelligence index22.9→19.5-15%
ScoreStep3 VL 10BStepFunIntelligence index3.8→7.7+103%
ScoreTiny Aya GlobalCohereIntelligence index1→4.8+380%
ScoreTri-21B-ThinkTrillion LabsIntelligence index6.6→9+36%
ScoreTri-21B-think PreviewTrillion LabsIntelligence index7.7→9.6+25%
ScoreTrinity Large ThinkingArcee AIIntelligence index13.1→10.9-17%

09-08

Sep 8, 20261 changes

AddedMiniCPM5-2BOpenBMBIntelligence index 15

09-06

Sep 6, 20266 changes

ScoreGemini 3.1 Pro PreviewGoogleIntelligence index38.3→36.7-4.2%
ScoreGemini 3.1 Pro PreviewGoogleCoding index68.7→68.8+0.1%
ScoreMistral Large 3MistralIntelligence index9.7→11.1+14%
ScoreMistral Large 3MistralCoding index20.2→20.1-0.5%
ScoreMuse Spark 1.3 (max)MetaIntelligence index52.7→53+0.6%
ScoreQwen3.8 27B (xhigh)AlibabaIntelligence index41.6→41.4-0.5%

09-05

Sep 5, 202620 changes

PriceCommand-R (Mar '24)CohereInput price$0.500→Metric withdrawn
PriceCommand-R (Mar '24)CohereOutput price$1.50→Metric withdrawn
PriceCommand-R+ (Apr '24)CohereInput price$3.00→Metric withdrawn
PriceCommand-R+ (Apr '24)CohereOutput price$15.00→Metric withdrawn
PriceGemma 4 31B (Non-reasoning)GoogleInput price$0.150→$0.140-6.7%
PriceGLM-5.1 (Reasoning)Z AIInput price$1.28→$1.20-6.3%
PriceGLM-5.1 (Reasoning)Z AIOutput price$4.07→$4.40+8.1%
Pricegpt-oss-120b (high)OpenAIOutput price$0.600→$0.590-1.7%
Pricegpt-oss-120b (low)OpenAIOutput price$0.590→$0.540-8.5%
PriceKimi K2.5 (Reasoning)KimiOutput price$3.00→$2.75-8.3%
PriceLlama 4 ScoutMetaInput price$0.180→$0.190+5.6%
PriceLlama 4 ScoutMetaOutput price$0.660→$0.680+3.0%
PriceMuse Glimmer (high)MetaInput price$0.320→$0.350+9.4%
PriceMuse Glimmer (high)MetaOutput price$1.35→$1.50+11%
PriceMuse Spark 1.3 (max)MetaInput priceFirst measured→$1.25
PriceMuse Spark 1.3 (max)MetaOutput priceFirst measured→$4.25
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAInput price$0.680→$0.600-12%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAOutput price$2.67→$2.60-2.6%
PriceNemotron 3.5 LightningNVIDIAInput price$0.070→$0.060-14%
PriceNemotron 3.5 LightningNVIDIAOutput price$0.220→$0.200-9.1%

09-04

Sep 4, 20269 changes

AddedGPT-6 Astra (high)OpenAIIntelligence index 60.3
AddedGPT-6 Astra (low)OpenAIIntelligence index 56.7
AddedGPT-6 Astra (max)OpenAIIntelligence index 61.2
AddedGPT-6 Astra (medium)OpenAIIntelligence index 59.2
AddedGPT-6 Astra (Non-reasoning)OpenAIIntelligence index 55.3
AddedGPT-6 Astra (xhigh)OpenAIIntelligence index 61
AddedK2 Horizon 375B A23BMBZUAI Institute of Foundation ModelsIntelligence index 47.3
AddedMuse Spark 1.3 (max)MetaIntelligence index 62.1
AddedMuse Spark 1.3 (xhigh)MetaIntelligence index 60.8

09-03

Sep 3, 20263 changes

AddedGemini 3.8 Flash (high)GoogleIntelligence index 58.7
AddedGemini 3.8 Flash (low)GoogleIntelligence index 51.7
AddedGemini 3.8 Flash (medium)GoogleIntelligence index 56.6

09-02

Sep 2, 202611 changes

AddedClaude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)AnthropicIntelligence index 62.5
AddedClaude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)AnthropicIntelligence index 58.1
AddedClaude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)AnthropicIntelligence index 65.7
AddedClaude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)AnthropicIntelligence index 60.5
AddedClaude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)AnthropicIntelligence index 64.8
AddedNVIDIA Nemotron 3 Nano 4BNVIDIAIntelligence index 8.6
AddedQuasar 438B (max, based on GLM-5.2)Multiverse ComputingIntelligence index 43
PriceGemma 3n E4B InstructGoogleInput price$0.060→Metric withdrawn
PriceGemma 3n E4B InstructGoogleOutput price$0.120→Metric withdrawn
PriceTrinity Large ThinkingArcee AIInput price$0.230→$0.250+8.7%
PriceTrinity Large ThinkingArcee AIOutput price$0.880→$0.900+2.3%

09-01

Sep 1, 20267 changes

AddedApodex 1.1ApodexIntelligence index 44
RemovedNVIDIA Nemotron 3 Nano 4BNVIDIA
PriceGLM-5.1 (Non-reasoning)Z AIInput price$1.39→$1.38-0.7%
PriceGLM-5.1 (Reasoning)Z AIInput price$1.38→$1.28-7.2%
PriceGLM-5.1 (Reasoning)Z AIOutput price$4.40→$4.07-7.5%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAInput price$0.600→$0.680+13%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAOutput price$2.75→$2.67-2.9%

08-28

Aug 28, 20263 changes

AddedAgnes 2.5 Pro BetaSapiens AIIntelligence index 49.1
AddedGranite 4.2 30BIBMIntelligence index 23.7
AddedQwen3.8-Flash-NextAlibabaIntelligence index 55.8

08-27

Aug 27, 20269 changes

AddedDeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeekIntelligence index 51.5
AddedGLM 5.3 FlashZ AIIntelligence index 57.5
AddedGranite 4.2 3BIBMIntelligence index 14.3
AddedGranite 4.2 8BIBMIntelligence index 19.6
AddedQwen3.8 27B (Non-reasoning)AlibabaIntelligence index 34.7
PriceDeepSeek V3.1 (Non-reasoning)DeepSeekInput price$0.560→$0.570+1.8%
PriceGLM-5.1 (Reasoning)Z AIInput price$1.39→$1.38-0.7%
PriceHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingInput priceFirst measured→$0.040
PriceHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingOutput priceFirst measured→$0.140

08-25

Aug 25, 202625 changes

AddedG9v3-39A5BAI9StarsIntelligence index 34
AddedGranite 4.0 1BIBMIntelligence index 1.6
AddedGrok 4.6 (low)SpaceXAIIntelligence index 51.7
AddedGrok 4.6 (medium)SpaceXAIIntelligence index 59
AddedGrok 4.6 (xhigh)SpaceXAIIntelligence index 60
AddedK2-V2 (high)MBZUAI Institute of Foundation ModelsIntelligence index 14.2
AddedLFM2.5-2.6BLiquid AIIntelligence index 11
AddedMotif 3 (Beta)Motif TechnologiesIntelligence index 45.3
RemovedG9v3-39A5BAI9Stars
PriceGPT-5.6 Sol (high)OpenAIInput price$5.00→$4.00-20%
PriceGPT-5.6 Sol (high)OpenAIOutput price$30.00→$20.00-33%
PriceGPT-5.6 Sol (low)OpenAIInput price$5.00→$4.00-20%
PriceGPT-5.6 Sol (low)OpenAIOutput price$30.00→$20.00-33%
PriceGPT-5.6 Sol (max)OpenAIInput price$5.00→$4.00-20%
PriceGPT-5.6 Sol (max)OpenAIOutput price$30.00→$20.00-33%
PriceGPT-5.6 Sol (medium)OpenAIInput price$5.00→$4.00-20%
PriceGPT-5.6 Sol (medium)OpenAIOutput price$30.00→$20.00-33%
PriceGPT-5.6 Sol (Non-reasoning)OpenAIInput price$5.00→$4.00-20%
PriceGPT-5.6 Sol (Non-reasoning)OpenAIOutput price$30.00→$20.00-33%
PriceGPT-5.6 Sol (xhigh)OpenAIInput price$5.00→$4.00-20%
PriceGPT-5.6 Sol (xhigh)OpenAIOutput price$30.00→$20.00-33%
PriceLing 2.6 FlashInclusionAIInput price$0.100→Metric withdrawn
PriceLing 2.6 FlashInclusionAIOutput price$0.300→Metric withdrawn
PriceQwen3.8 27B (xhigh)AlibabaInput price$0.450→$0.500+11%
PriceQwen3.8 27B (xhigh)AlibabaOutput price$3.20→$3.00-6.3%

08-24

Aug 24, 20269 changes

AddedQwen3.8 27B (low)AlibabaIntelligence index 42.9
AddedQwen3.8 27B (medium)AlibabaIntelligence index 44.5
RemovedGranite 4.0 1BIBM
RemovedK2-V2 (high)MBZUAI Institute of Foundation Models
RemovedMotif 3 (Beta)Motif Technologies
PriceLlama 4 MaverickMetaInput price$0.270→$0.260-3.7%
PriceLlama 4 MaverickMetaOutput price$0.850→$0.910+7.1%
PriceMagistral Medium 1.2MistralInput price$2.00→Metric withdrawn
PriceMagistral Medium 1.2MistralOutput price$5.00→Metric withdrawn

08-20

Aug 20, 202639 changes

AddedGLM-5.3 (max)Z AIIntelligence index 59.5
AddedQwen3.8 27B (xhigh)AlibabaIntelligence index 52
PriceDeepSeek R1 (Jan '25)DeepSeekInput price$1.68→$2.00+19%
PriceDeepSeek R1 (Jan '25)DeepSeekOutput price$4.70→$4.00-15%
PriceDeepSeek R1 0528 (May '25)DeepSeekOutput price$4.20→$3.00-29%
PriceDeepSeek V4 Flash 0420 (Non-reasoning)DeepSeekInput price$0.140→$0.090-36%
PriceDeepSeek V4 Flash 0420 (Non-reasoning)DeepSeekOutput price$0.280→$0.200-29%
PriceDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekInput price$0.140→$0.130-7.1%
PriceGemini 3.6 Flash (high)GoogleInput price$1.50→$0.750-50%
PriceGemini 3.6 Flash (high)GoogleOutput price$7.50→$3.75-50%
PriceGemma 4 26B A4B (Reasoning)GoogleInput price$0.130→$0.120-7.7%
PriceGemma 4 26B A4B (Reasoning)GoogleOutput price$0.400→$0.370-7.5%
PriceGLM-5.1 (Non-reasoning)Z AIInput price$1.38→$1.39+0.7%
PriceGLM-5.1 (Reasoning)Z AIInput price$1.38→$1.39+0.7%
PriceHy3TencentOutput price$0.560→$0.550-1.8%
PriceNemotron 3.5 LightningNVIDIAInput price$0.050→$0.070+40%
PriceNemotron 3.5 LightningNVIDIAOutput price$0.200→$0.220+10.0%
PriceQwen3 235B A22B 2507 (Reasoning)AlibabaInput price$0.700→$0.230-67%
PriceQwen3 235B A22B 2507 (Reasoning)AlibabaOutput price$8.40→$2.30-73%
PriceQwen3 235B A22B 2507 InstructAlibabaInput price$0.700→$0.230-67%
PriceQwen3 235B A22B 2507 InstructAlibabaOutput price$2.80→$0.920-67%
PriceQwen3 32B (Non-reasoning)AlibabaInput price$0.700→$0.160-77%
PriceQwen3 32B (Non-reasoning)AlibabaOutput price$2.80→$0.640-77%
PriceQwen3 32B (Reasoning)AlibabaInput price$0.700→$0.160-77%
PriceQwen3 32B (Reasoning)AlibabaOutput price$8.40→$0.640-92%
PriceQwen3 Next 80B A3B (Reasoning)AlibabaInput price$0.500→$0.150-70%
PriceQwen3 Next 80B A3B (Reasoning)AlibabaOutput price$6.00→$1.20-80%
PriceQwen3 Next 80B A3B InstructAlibabaInput price$0.500→$0.150-70%
PriceQwen3 Next 80B A3B InstructAlibabaOutput price$2.00→$1.20-40%
PriceQwen3 VL 235B A22B (Reasoning)AlibabaInput price$0.700→$0.400-43%
PriceQwen3 VL 235B A22B (Reasoning)AlibabaOutput price$8.40→$4.00-52%
PriceQwen3 VL 235B A22B InstructAlibabaInput price$0.700→$0.400-43%
PriceQwen3 VL 235B A22B InstructAlibabaOutput price$2.80→$1.60-43%
PriceQwen3 VL 32B (Reasoning)AlibabaInput price$0.700→$0.160-77%
PriceQwen3 VL 32B (Reasoning)AlibabaOutput price$8.40→$0.640-92%
PriceQwen3 VL 32B InstructAlibabaInput price$0.700→$0.160-77%
PriceQwen3 VL 32B InstructAlibabaOutput price$2.80→$0.640-77%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekIntelligence index31.1→31.4+1.0%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekAgentic indexFirst measured→19.9

08-16

Aug 16, 20263 changes

ScoreLongCat 2.0LongCatIntelligence index34.3→34-0.9%
ScoreLongCat 2.0LongCatAgentic indexFirst measured→22
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaAgentic indexFirst measured→19.2

08-15

Aug 15, 202610 changes

PriceDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekInput price$0.140→$0.440+214%
PriceDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekOutput price$0.280→$1.32+371%
ScoreKAT Coder Pro V2KwaiKATIntelligence index34.5→33.7-2.3%
ScoreKAT Coder Pro V2KwaiKATAgentic indexFirst measured→15.1
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIIntelligence index42.1→41.7-1.0%
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIAgentic indexFirst measured→31.2
ScoreRing-2.6-1TInclusionAIIntelligence index31.3→31.7+1.3%
ScoreRing-2.6-1TInclusionAIAgentic indexFirst measured→20.1
ScoreTrinity Large ThinkingArcee AIIntelligence index18.4→18.7+1.6%
ScoreTrinity Large ThinkingArcee AIAgentic indexFirst measured→3.7

08-14

Aug 14, 20268 changes

AddedGemini 3.7 Flash (high)GoogleIntelligence index 56
AddedGemini 3.7 Flash (low)GoogleIntelligence index 50.9
AddedGemini 3.7 Flash (medium)GoogleIntelligence index 53.4
AddedQwen3.8 2.4T A95BAlibabaIntelligence index 57.7
RemovedClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 5 Fallback)Anthropic
PriceDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekInput price$0.430→$1.32+207%
PriceDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekOutput price$0.870→$3.96+355%
ScoreDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekIntelligence index53→53.2+0.4%

08-13

Aug 13, 202619 changes

AddedA.X-K2SK TelecomIntelligence index 35
AddedClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 5 Fallback)Anthropic
AddedDeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeekIntelligence index 53
AddedGrok 4.6 (high)SpaceXAIIntelligence index 60.9
AddedK-EXAONE 2.0 0803LG AI ResearchIntelligence index 31
AddedMotif 3Motif TechnologiesIntelligence index 47.4
AddedSolar Open2 250BUpstageIntelligence index 37.4
AddedSolar Pro 4UpstageIntelligence index 41.6
PriceClaude 3 SonnetAnthropicInput price$3.00→Metric withdrawn
PriceClaude 3 SonnetAnthropicOutput price$15.00→Metric withdrawn
Pricegpt-oss-120b (high)OpenAIOutput price$0.590→$0.600+1.7%
PriceJamba 1.6 LargeAI21 LabsInput price$2.00→Metric withdrawn
PriceJamba 1.6 LargeAI21 LabsOutput price$8.00→Metric withdrawn
PriceJamba 1.6 MiniAI21 LabsInput price$0.200→Metric withdrawn
PriceJamba 1.6 MiniAI21 LabsOutput price$0.400→Metric withdrawn
PriceJamba 1.7 LargeAI21 LabsInput price$2.00→Metric withdrawn
PriceJamba 1.7 LargeAI21 LabsOutput price$8.00→Metric withdrawn
PriceMuse Glimmer (high)MetaInput priceFirst measured→$0.320
PriceMuse Glimmer (high)MetaOutput priceFirst measured→$1.35

08-12

Aug 12, 202695 changes

AddedMuse Glimmer (high)MetaIntelligence index 35.1
AddedNemotron 3.5 LightningNVIDIAIntelligence index 23.6
PriceDeepSeek R1 Distill Llama 70BDeepSeekOutput price$1.05→$1.10+4.8%
PriceGLM-5.2 (max)Z AIInput price$1.35→$1.40+3.7%
PriceGLM-5.2 (max)Z AIOutput price$4.29→$4.40+2.6%
PriceGLM-5.2 (Non-reasoning)Z AIInput price$1.35→$1.40+3.7%
PriceGLM-5.2 (Non-reasoning)Z AIOutput price$4.25→$4.40+3.5%
Pricegpt-oss-120b (low)OpenAIOutput price$0.540→$0.590+9.3%
PriceHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingInput price$0.040→Metric withdrawn
PriceHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingOutput price$0.140→Metric withdrawn
PriceLlama 3.1 Instruct 8BMetaInput price$0.050→$0.020-60%
PriceLlama 3.1 Instruct 8BMetaOutput price$0.080→$0.050-38%
PriceQwen3.5 9B (Non-reasoning)AlibabaInput priceFirst measured→$0.170
PriceQwen3.5 9B (Non-reasoning)AlibabaOutput priceFirst measured→$0.250
PriceQwen3.6 35B A3B (Reasoning)AlibabaInput price$0.250→$0.380+52%
PriceQwen3.6 35B A3B (Reasoning)AlibabaOutput price$1.49→$2.25+51%
ScoreCeleris-1CelerisIntelligence index12.5→12.4-0.8%
ScoreCeleris-1CelerisAgentic index2.8→2.4-14%
ScoreClaude 3.5 HaikuAnthropicIntelligence index12.4→12.2-1.6%
ScoreClaude 3.5 HaikuAnthropicAgentic index1.9→Metric withdrawn
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekIntelligence index30.6→31.1+1.6%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekAgentic index18.1→Metric withdrawn
ScoreDeepSeek V3.2 (Reasoning)DeepSeekIntelligence index32.6→32.8+0.6%
ScoreDeepSeek V3.2 (Reasoning)DeepSeekAgentic index18.3→Metric withdrawn
ScoreDiffusionGemma 26B A4BGoogleIntelligence index13.8→13.5-2.2%
ScoreDiffusionGemma 26B A4BGoogleAgentic index2.2→Metric withdrawn
ScoreEXAONE 4.5 33BLG AI ResearchAgentic index9.7→Metric withdrawn
ScoreG9v3-39A5BAI9StarsIntelligence index31.4→31.6+0.6%
ScoreG9v3-39A5BAI9StarsAgentic index28.7→Metric withdrawn
ScoreG9v3-3BAI9StarsIntelligence index16.5→16.2-1.8%
ScoreG9v3-3BAI9StarsAgentic index14→Metric withdrawn
ScoreGemma 4 12B (Reasoning)GoogleIntelligence index22.3→22.2-0.4%
ScoreGemma 4 12B (Reasoning)GoogleAgentic index7.9→Metric withdrawn
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index9.8→9.5-3.1%
ScoreGemma 4 E2B (Reasoning)GoogleAgentic index1.5→Metric withdrawn
ScoreGemma 4 E4B (Reasoning)GoogleIntelligence index12.4→12.2-1.6%
ScoreGemma 4 E4B (Reasoning)GoogleAgentic index1.8→Metric withdrawn
ScoreGranite 4.1 30BIBMIntelligence index8.9→8.7-2.2%
ScoreGranite 4.1 30BIBMAgentic index1.3→Metric withdrawn
ScoreGranite 4.1 3BIBMIntelligence index4.7→4.4-6.4%
ScoreGranite 4.1 3BIBMAgentic index0.4→Metric withdrawn
ScoreJT-4.1 Flash 236B A21BChina MobileIntelligence index39.8→39.9+0.3%
ScoreJT-4.1 Flash 236B A21BChina MobileAgentic index34.9→Metric withdrawn
ScoreK-EXAONE (Reasoning)LG AI ResearchIntelligence index22.4→22.5+0.4%
ScoreK-EXAONE (Reasoning)LG AI ResearchAgentic index8→Metric withdrawn
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsIntelligence index17.2→17.4+1.2%
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsAgentic index1.8→Metric withdrawn
ScoreKAT Coder Pro V2KwaiKATIntelligence index33.9→34.5+1.8%
ScoreKAT Coder Pro V2KwaiKATAgentic index15.5→Metric withdrawn
ScoreLing 3.0 TinyInclusionAIIntelligence index24.3→24.5+0.8%
ScoreLing 3.0 TinyInclusionAIAgentic index15.4→16+3.9%
ScoreLlama 3.1 Instruct 8BMetaIntelligence index7.8→7.4-5.1%
ScoreLlama 3.1 Instruct 8BMetaAgentic index0.5→Metric withdrawn
ScoreLlama 3.3 Instruct 70BMetaAgentic index0.3→Metric withdrawn
ScoreLongCat 2.0LongCatIntelligence index33.9→34.3+1.2%
ScoreLongCat 2.0LongCatAgentic index21.8→Metric withdrawn
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiIntelligence index25→25.1+0.4%
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiAgentic index12→Metric withdrawn
ScoreMiniCPM-V 4.6 1.3BOpenBMBIntelligence index4.3→3.8-12%
ScoreMiniCPM-V 4.6 1.3BOpenBMBAgentic index1.4→Metric withdrawn
ScoreMotif 3 (Beta)Motif TechnologiesIntelligence index44.9→45.3+0.9%
ScoreMotif 3 (Beta)Motif TechnologiesAgentic index34.9→Metric withdrawn
ScoreMuse SparkMetaIntelligence index43.5→44.3+1.8%
ScoreMuse SparkMetaAgentic index28.7→Metric withdrawn
ScoreNanbeige4.1-3BNanbeigeIntelligence index11.2→11-1.8%
ScoreNanbeige4.1-3BNanbeigeAgentic index0→Metric withdrawn
ScoreNemotron Cascade 2 30B A3BNVIDIAIntelligence index18→17.8-1.1%
ScoreNemotron Cascade 2 30B A3BNVIDIAAgentic index3.7→Metric withdrawn
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIIntelligence index41.7→42.1+1.0%
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIAgentic index31→Metric withdrawn
ScoreNova 2.0 Lite (high)AmazonIntelligence index18.4→20.8+13%
ScoreNova 2.0 Lite (high)AmazonAgentic index3.1→Metric withdrawn
ScoreNova 2.0 Pro Preview (low)AmazonIntelligence index20→19.8-1.0%
ScoreNova 2.0 Pro Preview (low)AmazonAgentic index6.4→Metric withdrawn
ScoreNova 2.0 Pro Preview (medium)AmazonIntelligence index22→22.1+0.5%
ScoreNova 2.0 Pro Preview (medium)AmazonAgentic index7→Metric withdrawn
ScoreNova 2.0 Pro Preview (Non-reasoning)AmazonAgentic index2.9→Metric withdrawn
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAIntelligence index8.8→8.6-2.3%
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAAgentic index2.3→Metric withdrawn
ScorePhi-4 Mini InstructMicrosoftIntelligence index6.2→5.7-8.1%
ScorePhi-4 Mini InstructMicrosoftAgentic index0.3→Metric withdrawn
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaIntelligence index3.5→2.9-17%
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaAgentic index0.4→Metric withdrawn
ScoreQwen3.5 0.8B (Reasoning)AlibabaIntelligence index5→5.2+4.0%
ScoreQwen3.5 0.8B (Reasoning)AlibabaAgentic index5.7→Metric withdrawn
ScoreQwen3.5 2B (Non-reasoning)AlibabaIntelligence index5.7→5.3-7.0%
ScoreQwen3.5 2B (Non-reasoning)AlibabaAgentic index0.8→Metric withdrawn
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index7.5→7.4-1.3%
ScoreQwen3.5 2B (Reasoning)AlibabaAgentic index0.8→Metric withdrawn
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaIntelligence index24.5→24.6+0.4%
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaAgentic index19.1→Metric withdrawn
ScoreRing-2.6-1TInclusionAIIntelligence index31.1→31.3+0.6%
ScoreRing-2.6-1TInclusionAIAgentic index18.9→Metric withdrawn
ScoreTrinity Large ThinkingArcee AIIntelligence index18.6→18.4-1.1%
ScoreTrinity Large ThinkingArcee AIAgentic index3.7→Metric withdrawn

08-08

Aug 8, 202685 changes

PriceGLM-5.2 (max)Z AIInput price$1.30→$1.35+3.8%
PriceGLM-5.2 (max)Z AIOutput price$4.18→$4.29+2.6%
Pricegpt-oss-120b (low)OpenAIOutput price$0.590→$0.540-8.5%
ScoreCeleris-1CelerisIntelligence index12.4→12.5+0.8%
ScoreCeleris-1CelerisAgentic index2.4→2.8+17%
ScoreClaude 3.5 HaikuAnthropicIntelligence index12.2→12.4+1.6%
ScoreClaude 3.5 HaikuAnthropicAgentic indexFirst measured→1.9
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekIntelligence index31.1→30.6-1.6%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekAgentic indexFirst measured→18.1
ScoreDeepSeek V3.2 (Reasoning)DeepSeekIntelligence index32.8→32.6-0.6%
ScoreDeepSeek V3.2 (Reasoning)DeepSeekAgentic indexFirst measured→18.3
ScoreDiffusionGemma 26B A4BGoogleIntelligence index13.5→13.8+2.2%
ScoreDiffusionGemma 26B A4BGoogleAgentic indexFirst measured→2.2
ScoreEXAONE 4.5 33BLG AI ResearchAgentic indexFirst measured→9.7
ScoreG9v3-39A5BAI9StarsIntelligence index31.6→31.4-0.6%
ScoreG9v3-39A5BAI9StarsAgentic indexFirst measured→28.7
ScoreG9v3-3BAI9StarsIntelligence index16.2→16.5+1.9%
ScoreG9v3-3BAI9StarsAgentic indexFirst measured→14
ScoreGemma 4 12B (Reasoning)GoogleIntelligence index22.2→22.3+0.5%
ScoreGemma 4 12B (Reasoning)GoogleAgentic indexFirst measured→7.9
ScoreGemma 4 31B (Reasoning)GoogleIntelligence index30→29.7-1.0%
ScoreGemma 4 31B (Reasoning)GoogleAgentic indexFirst measured→14.4
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index9.5→9.8+3.2%
ScoreGemma 4 E2B (Reasoning)GoogleAgentic indexFirst measured→1.5
ScoreGemma 4 E4B (Reasoning)GoogleIntelligence index12.2→12.4+1.6%
ScoreGemma 4 E4B (Reasoning)GoogleAgentic indexFirst measured→1.8
ScoreGranite 4.1 30BIBMIntelligence index8.7→8.9+2.3%
ScoreGranite 4.1 30BIBMAgentic indexFirst measured→1.3
ScoreGranite 4.1 3BIBMIntelligence index4.4→4.7+6.8%
ScoreGranite 4.1 3BIBMAgentic indexFirst measured→0.4
ScoreJT-4.1 Flash 236B A21BChina MobileIntelligence index39.9→39.8-0.3%
ScoreJT-4.1 Flash 236B A21BChina MobileAgentic indexFirst measured→34.9
ScoreK-EXAONE (Reasoning)LG AI ResearchIntelligence index22.5→22.4-0.4%
ScoreK-EXAONE (Reasoning)LG AI ResearchAgentic indexFirst measured→8
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsIntelligence index17.4→17.2-1.1%
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsAgentic indexFirst measured→1.8
ScoreKAT Coder Pro V2KwaiKATIntelligence index34.5→33.9-1.7%
ScoreKAT Coder Pro V2KwaiKATAgentic indexFirst measured→15.5
ScoreLing 3.0 TinyInclusionAIIntelligence index23.9→24.3+1.7%
ScoreLing 3.0 TinyInclusionAICoding index25.8→26.5+2.7%
ScoreLing 3.0 TinyInclusionAIAgentic indexFirst measured→15.4
ScoreLlama 3.1 Instruct 8BMetaIntelligence index7.4→7.8+5.4%
ScoreLlama 3.1 Instruct 8BMetaAgentic indexFirst measured→0.5
ScoreLlama 3.3 Instruct 70BMetaAgentic indexFirst measured→0.3
ScoreLongCat 2.0LongCatIntelligence index34.3→33.9-1.2%
ScoreLongCat 2.0LongCatAgentic indexFirst measured→21.8
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiIntelligence index25.1→25-0.4%
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiAgentic indexFirst measured→12
ScoreMiniCPM-V 4.6 1.3BOpenBMBIntelligence index3.8→4.3+13%
ScoreMiniCPM-V 4.6 1.3BOpenBMBAgentic indexFirst measured→1.4
ScoreMotif 3 (Beta)Motif TechnologiesIntelligence index45.3→44.9-0.9%
ScoreMotif 3 (Beta)Motif TechnologiesAgentic indexFirst measured→34.9
ScoreMuse SparkMetaIntelligence index44.3→43.5-1.8%
ScoreMuse SparkMetaAgentic indexFirst measured→28.7
ScoreNanbeige4.1-3BNanbeigeIntelligence index11→11.2+1.8%
ScoreNanbeige4.1-3BNanbeigeAgentic indexFirst measured→0
ScoreNemotron Cascade 2 30B A3BNVIDIAIntelligence index17.8→18+1.1%
ScoreNemotron Cascade 2 30B A3BNVIDIAAgentic indexFirst measured→3.7
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIIntelligence index42.1→41.7-1.0%
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIAgentic indexFirst measured→31
ScoreNova 2.0 Lite (high)AmazonIntelligence index20.8→18.4-12%
ScoreNova 2.0 Lite (high)AmazonAgentic indexFirst measured→3.1
ScoreNova 2.0 Pro Preview (low)AmazonIntelligence index19.8→20+1.0%
ScoreNova 2.0 Pro Preview (low)AmazonAgentic indexFirst measured→6.4
ScoreNova 2.0 Pro Preview (medium)AmazonIntelligence index22.1→22-0.5%
ScoreNova 2.0 Pro Preview (medium)AmazonAgentic indexFirst measured→7
ScoreNova 2.0 Pro Preview (Non-reasoning)AmazonAgentic indexFirst measured→2.9
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAIntelligence index8.6→8.8+2.3%
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAAgentic indexFirst measured→2.3
ScorePhi-4 Mini InstructMicrosoftIntelligence index5.7→6.2+8.8%
ScorePhi-4 Mini InstructMicrosoftAgentic indexFirst measured→0.3
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaIntelligence index2.9→3.5+21%
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaAgentic indexFirst measured→0.4
ScoreQwen3.5 0.8B (Reasoning)AlibabaIntelligence index5.2→5-3.8%
ScoreQwen3.5 0.8B (Reasoning)AlibabaAgentic indexFirst measured→5.7
ScoreQwen3.5 2B (Non-reasoning)AlibabaIntelligence index5.3→5.7+7.5%
ScoreQwen3.5 2B (Non-reasoning)AlibabaAgentic indexFirst measured→0.8
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index7.4→7.5+1.4%
ScoreQwen3.5 2B (Reasoning)AlibabaAgentic indexFirst measured→0.8
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaIntelligence index24.6→24.5-0.4%
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaAgentic indexFirst measured→19.1
ScoreRing-2.6-1TInclusionAIIntelligence index31.3→31.1-0.6%
ScoreRing-2.6-1TInclusionAIAgentic indexFirst measured→18.9
ScoreTrinity Large ThinkingArcee AIIntelligence index18.4→18.6+1.1%
ScoreTrinity Large ThinkingArcee AIAgentic indexFirst measured→3.7

08-07

Aug 7, 2026692 changes

AddedLing 3.0 TinyInclusionAIIntelligence index 23.9
AddedLing 3.0 FlashInclusionAIIntelligence index 37.8
AddedMuse Spark 1.2 (xhigh)MetaIntelligence index 56.8
AddedQwen3.8 MaxAlibabaIntelligence index 58.1
PriceDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekInput price$0.140→$0.130-7.1%
PriceGLM-5.2 (max)Z AIInput price$1.35→$1.30-3.7%
PriceGLM-5.2 (max)Z AIOutput price$4.29→$4.18-2.6%
Pricegpt-oss-120b (high)OpenAIOutput price$0.600→$0.590-1.7%
Pricegpt-oss-20b (high)OpenAIInput price$0.070→$0.060-14%
Pricegpt-oss-20b (high)OpenAIOutput price$0.200→$0.190-5.0%
Pricegpt-oss-20b (low)OpenAIOutput price$0.230→$0.200-13%
PriceLlama 3.1 Instruct 405BMetaInput price$2.50→Metric withdrawn
PriceLlama 3.1 Instruct 405BMetaOutput price$10.00→Metric withdrawn
PriceLlama 3.1 Instruct 8BMetaInput price$0.080→$0.050-38%
PriceLlama 3.1 Instruct 8BMetaOutput price$0.090→$0.080-11%
PriceLlama 3.3 Instruct 70BMetaInput price$0.600→$0.660+10%
PriceLlama 4 MaverickMetaOutput price$0.910→$0.850-6.6%
ScoreAgnes 2.5 Pro AlphaSapiens AIIntelligence index38.8→39.7+2.3%
ScoreAgnes 2.5 Pro AlphaSapiens AIAgentic index25.2→25.4+0.8%
ScoreApertus 70B InstructSwiss AI InitiativeIntelligence index2.4→2-17%
ScoreApriel-v1.5-15B-ThinkerServiceNowIntelligence index21.2→21.6+1.9%
ScoreApriel-v1.6-15B-ThinkerServiceNowIntelligence index20.5→20.8+1.5%
ScoreArctic InstructSnowflakeIntelligence index3.4→3-12%
ScoreCeleris-1CelerisIntelligence index11.8→12.4+5.1%
ScoreCeleris-1CelerisAgentic index2.8→2.4-14%
ScoreClaude 2.0AnthropicIntelligence index3.6→3.3-8.3%
ScoreClaude 2.1AnthropicIntelligence index3.9→3.5-10%
ScoreClaude 3 HaikuAnthropicIntelligence index3.9→3.5-10%
ScoreClaude 3 SonnetAnthropicIntelligence index4.7→4.4-6.4%
ScoreClaude 3.5 HaikuAnthropicIntelligence index12.3→12.2-0.8%
ScoreClaude 3.5 HaikuAnthropicAgentic index1.9→Metric withdrawn
ScoreClaude 3.5 Sonnet (June '24)AnthropicIntelligence index8.3→8.1-2.4%
ScoreClaude 3.5 Sonnet (Oct '24)AnthropicIntelligence index9.9→9.8-1.0%
ScoreClaude 3.7 Sonnet (Non-reasoning)AnthropicIntelligence index23.5→23.9+1.7%
ScoreClaude 3.7 Sonnet (Reasoning)AnthropicIntelligence index27.1→27.6+1.8%
ScoreClaude 4 Opus (Non-reasoning)AnthropicIntelligence index25.5→26+2.0%
ScoreClaude 4 Opus (Reasoning)AnthropicIntelligence index31→31.7+2.3%
ScoreClaude 4 Sonnet (Non-reasoning)AnthropicIntelligence index25.5→26+2.0%
ScoreClaude 4 Sonnet (Reasoning)AnthropicIntelligence index28.9→29.8+3.1%
ScoreClaude 4 Sonnet (Reasoning)AnthropicAgentic index16.6→17.6+6.0%
ScoreClaude 4.1 Opus (Non-reasoning)AnthropicIntelligence index28.2→28.8+2.1%
ScoreClaude 4.1 Opus (Reasoning)AnthropicIntelligence index33.7→34.5+2.4%
ScoreClaude 4.5 Haiku (Non-reasoning)AnthropicIntelligence index23.7→24.1+1.7%
ScoreClaude 4.5 Haiku (Reasoning)AnthropicIntelligence index29.6→29.9+1.0%
ScoreClaude 4.5 Haiku (Reasoning)AnthropicAgentic index16.4→16.5+0.6%
ScoreClaude 4.5 Sonnet (Non-reasoning)AnthropicIntelligence index29.3→29.9+2.0%
ScoreClaude 4.5 Sonnet (Reasoning)AnthropicIntelligence index36.4→37.4+2.7%
ScoreClaude 4.5 Sonnet (Reasoning)AnthropicAgentic index24.6→26.4+7.3%
ScoreClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)AnthropicIntelligence index59.9→62.1+3.7%
ScoreClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)AnthropicAgentic index52.8→56.6+7.2%
ScoreClaude InstantAnthropicIntelligence index2.1→1.7-19%
ScoreClaude Opus 4.5 (Non-reasoning)AnthropicIntelligence index34.7→35.6+2.6%
ScoreClaude Opus 4.5 (Reasoning)AnthropicIntelligence index40.8→41.9+2.7%
ScoreClaude Opus 4.6 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index43.7→44.9+2.7%
ScoreClaude Opus 4.6 (Non-reasoning, High Effort)AnthropicIntelligence index37.8→38.8+2.6%
ScoreClaude Opus 4.7 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index53.5→55+2.8%
ScoreClaude Opus 4.7 (Adaptive Reasoning, Max Effort)AnthropicAgentic index44.4→46.3+4.3%
ScoreClaude Opus 4.7 (Non-reasoning, High Effort)AnthropicIntelligence index42.7→43.9+2.8%
ScoreClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index55.7→57.3+2.9%
ScoreClaude Opus 4.8 (Adaptive Reasoning, Max Effort)AnthropicAgentic index47.2→49.4+4.7%
ScoreClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicIntelligence index58.9→61.5+4.4%
ScoreClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicAgentic index52.1→56.1+7.7%
ScoreClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicIntelligence index50.6→52.5+3.8%
ScoreClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicAgentic index39.8→42.1+5.8%
ScoreClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index60.7→63.1+4.0%
ScoreClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicAgentic index55.3→59.2+7.1%
ScoreClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicIntelligence index56.3→58.6+4.1%
ScoreClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicAgentic index47.1→50.4+7.0%
ScoreClaude Opus 5 (Adaptive Reasoning, Xhigh Effort)AnthropicIntelligence index60.1→62.5+4.0%
ScoreClaude Opus 5 (Adaptive Reasoning, Xhigh Effort)AnthropicAgentic index54.5→58.4+7.2%
ScoreClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index47.2→48.4+2.5%
ScoreClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort)AnthropicAgentic index40.8→42.1+3.2%
ScoreClaude Sonnet 4.6 (Non-reasoning, High Effort)AnthropicIntelligence index35.9→36.8+2.5%
ScoreClaude Sonnet 4.6 (Non-reasoning, Low Effort)AnthropicIntelligence index34.3→35.1+2.3%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index53.4→55.3+3.6%
ScoreClaude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicAgentic index46.7→49.7+6.4%
ScoreClaude Sonnet 5 (Non-reasoning, High Effort)AnthropicIntelligence index41.7→42.6+2.2%
ScoreClaude Sonnet 5 (Non-reasoning, High Effort)AnthropicAgentic index33.7→34.2+1.5%
ScoreCommand ACohereIntelligence index7.7→7.5-2.6%
ScoreCommand A+CohereIntelligence index22.5→22.8+1.3%
ScoreCommand-R (Mar '24)CohereIntelligence index2.1→1.7-19%
ScoreCommand-R+ (Apr '24)CohereIntelligence index3→2.6-13%
ScoreDBRX InstructDatabricksIntelligence index3→2.6-13%
ScoreDeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous ResearchIntelligence index2.3→1.9-17%
ScoreDeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous ResearchIntelligence index5.3→5-5.7%
ScoreDeepSeek Coder V2 Lite InstructDeepSeekIntelligence index3.1→2.7-13%
ScoreDeepSeek LLM 67B Chat (V1)DeepSeekIntelligence index3→2.6-13%
ScoreDeepSeek R1 (Jan '25)DeepSeekIntelligence index18.5→18.6+0.5%
ScoreDeepSeek R1 0528 (May '25)DeepSeekIntelligence index20.1→20.4+1.5%
ScoreDeepSeek R1 0528 Qwen3 8BDeepSeekIntelligence index10.4→10.3-1.0%
ScoreDeepSeek R1 Distill Llama 70BDeepSeekIntelligence index9.9→9.8-1.0%
ScoreDeepSeek R1 Distill Llama 8BDeepSeekIntelligence index6.4→6.2-3.1%
ScoreDeepSeek R1 Distill Qwen 1.5BDeepSeekIntelligence index3.7→3.3-11%
ScoreDeepSeek R1 Distill Qwen 14BDeepSeekIntelligence index9.8→9.7-1.0%
ScoreDeepSeek V3 0324DeepSeekIntelligence index15.4→15.2-1.3%
ScoreDeepSeek V3 0324DeepSeekAgentic index1.5→1.6+6.7%
ScoreDeepSeek V3.1 (Non-reasoning)DeepSeekIntelligence index21→21.4+1.9%
ScoreDeepSeek V3.1 (Reasoning)DeepSeekIntelligence index20.7→21+1.4%
ScoreDeepSeek V3.1 Terminus (Non-reasoning)DeepSeekIntelligence index21.4→21.7+1.4%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekIntelligence index30.4→31.1+2.3%
ScoreDeepSeek V3.1 Terminus (Reasoning)DeepSeekAgentic index18.1→Metric withdrawn
ScoreDeepSeek V3.2 (Non-reasoning)DeepSeekIntelligence index24.7→25.1+1.6%
ScoreDeepSeek V3.2 (Reasoning)DeepSeekIntelligence index32→32.8+2.5%
ScoreDeepSeek V3.2 (Reasoning)DeepSeekAgentic index18.3→Metric withdrawn
ScoreDeepSeek V3.2 Exp (Non-reasoning)DeepSeekIntelligence index21.3→21.7+1.9%
ScoreDeepSeek V3.2 Exp (Reasoning)DeepSeekIntelligence index25.4→25.9+2.0%
ScoreDeepSeek V3.2 SpecialeDeepSeekIntelligence index22.2→22.6+1.8%
ScoreDeepSeek V4 Flash 0420 (Non-reasoning)DeepSeekIntelligence index28.7→29.3+2.1%
ScoreDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekIntelligence index37.5→39+4.0%
ScoreDeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeekAgentic index28.2→30.3+7.4%
ScoreDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekIntelligence index40.3→42.1+4.5%
ScoreDeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeekAgentic index31.1→33.7+8.4%
ScoreDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekIntelligence index49.9→51.8+3.8%
ScoreDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekAgentic index45.7→48.4+5.9%
ScoreDeepSeek V4 Pro 0424 (Non-reasoning)DeepSeekIntelligence index31.2→31.9+2.2%
ScoreDeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeekIntelligence index43.1→43.7+1.4%
ScoreDeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeekAgentic index34.4→35.3+2.6%
ScoreDeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeekIntelligence index44.3→45.3+2.3%
ScoreDeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeekAgentic index36.4→37.8+3.8%
ScoreDeepSeek-Coder-V2DeepSeekIntelligence index5.1→4.7-7.8%
ScoreDeepSeek-V2-ChatDeepSeekIntelligence index3.6→3.3-8.3%
ScoreDeepSeek-V2.5DeepSeekIntelligence index6.6→6.4-3.0%
ScoreDeepSeek-V2.5 (Dec '24)DeepSeekIntelligence index6.8→6.5-4.4%
ScoreDevstral Small (Jul '25)MistralIntelligence index9.3→9.1-2.2%
ScoreDevstral Small 2MistralIntelligence index17.4→17.7+1.7%
ScoreDevstral Small 2MistralAgentic index10.5→10.8+2.9%
ScoreDiffusionGemma 26B A4BGoogleAgentic index2.2→Metric withdrawn
ScoreDoubao Seed CodeByteDance SeedIntelligence index26→26.5+1.9%
ScoreERNIE 4.5 300B A47BBaiduIntelligence index9→8.9-1.1%
ScoreERNIE 5.0 Thinking PreviewBaiduIntelligence index21.9→22.3+1.8%
ScoreExaone 4.0 1.2B (Non-reasoning)LG AI ResearchIntelligence index2.8→2.4-14%
ScoreExaone 4.0 1.2B (Reasoning)LG AI ResearchIntelligence index2.9→2.5-14%
ScoreEXAONE 4.0 32B (Non-reasoning)LG AI ResearchIntelligence index6→5.7-5.0%
ScoreEXAONE 4.0 32B (Reasoning)LG AI ResearchIntelligence index10.6→10.5-0.9%
ScoreEXAONE 4.5 33BLG AI ResearchIntelligence index20.2→20.5+1.5%
ScoreEXAONE 4.5 33BLG AI ResearchAgentic index9.7→Metric withdrawn
ScoreFalcon-H1R-7BTII UAEIntelligence index9.8→9.7-1.0%
ScoreG9v3-39A5BAI9StarsIntelligence index30.9→31.6+2.3%
ScoreG9v3-39A5BAI9StarsAgentic index28.7→Metric withdrawn
ScoreG9v3-3BAI9StarsIntelligence index16.1→16.2+0.6%
ScoreG9v3-3BAI9StarsAgentic index14→Metric withdrawn
ScoreGemini 1.0 ProGoogleIntelligence index3.1→2.7-13%
ScoreGemini 1.0 UltraGoogleIntelligence index4.6→4.3-6.5%
ScoreGemini 1.5 Flash (May '24)GoogleIntelligence index4.9→4.6-6.1%
ScoreGemini 1.5 Flash (Sep '24)GoogleIntelligence index8→7.8-2.5%
ScoreGemini 1.5 Flash-8BGoogleIntelligence index5.5→5.2-5.5%
ScoreGemini 1.5 Pro (May '24)GoogleIntelligence index6.3→6.1-3.2%
ScoreGemini 1.5 Pro (Sep '24)GoogleIntelligence index10→9.9-1.0%
ScoreGemini 2.0 Flash (experimental)GoogleIntelligence index10.7→10.6-0.9%
ScoreGemini 2.0 Flash (Feb '25)GoogleIntelligence index12.3→12.2-0.8%
ScoreGemini 2.0 Flash Thinking Experimental (Dec '24)GoogleIntelligence index6.6→6.4-3.0%
ScoreGemini 2.0 Flash-Lite (Feb '25)GoogleIntelligence index8.8→8.6-2.3%
ScoreGemini 2.0 Flash-Lite (Preview)GoogleIntelligence index8.6→8.4-2.3%
ScoreGemini 2.5 Flash (Non-reasoning)GoogleIntelligence index14.1→14.2+0.7%
ScoreGemini 2.5 Flash (Reasoning)GoogleIntelligence index20.1→20.3+1.0%
ScoreGemini 2.5 Flash Preview (Non-reasoning)GoogleIntelligence index11.7→11.6-0.9%
ScoreGemini 2.5 Flash Preview (Reasoning)GoogleIntelligence index17.5→17.7+1.1%
ScoreGemini 2.5 Flash Preview (Sep '25) (Non-reasoning)GoogleIntelligence index18.8→19.1+1.6%
ScoreGemini 2.5 Flash Preview (Sep '25) (Reasoning)GoogleIntelligence index23.8→24.2+1.7%
ScoreGemini 2.5 Flash-Lite (Non-reasoning)GoogleIntelligence index6.9→6.7-2.9%
ScoreGemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)GoogleIntelligence index15.1→15.2+0.7%
ScoreGemini 2.5 ProGoogleIntelligence index25.8→25.9+0.4%
ScoreGemini 2.5 ProGoogleAgentic index7.1→7.2+1.4%
ScoreGemini 2.5 Pro Preview (Mar' 25)GoogleIntelligence index23→23.4+1.7%
ScoreGemini 2.5 Pro Preview (May' 25)GoogleIntelligence index22.3→22.7+1.8%
ScoreGemini 3 Flash Preview (Non-reasoning)GoogleIntelligence index27.4→27.9+1.8%
ScoreGemini 3 Flash Preview (Reasoning)GoogleIntelligence index37.8→38.7+2.4%
ScoreGemini 3 Pro Preview (high)GoogleIntelligence index39.6→40.6+2.5%
ScoreGemini 3 Pro Preview (low)GoogleIntelligence index33.1→33.9+2.4%
ScoreGemini 3.1 Flash-LiteGoogleIntelligence index25→25.6+2.4%
ScoreGemini 3.1 Flash-LiteGoogleAgentic index6.2→6.5+4.8%
ScoreGemini 3.1 Pro PreviewGoogleIntelligence index46.5→47.7+2.6%
ScoreGemini 3.1 Pro PreviewGoogleAgentic index21.4→23+7.5%
ScoreGemini 3.5 Flash (high)GoogleIntelligence index50.2→52+3.6%
ScoreGemini 3.5 Flash (high)GoogleAgentic index37.4→39.7+6.1%
ScoreGemini 3.5 Flash (medium)GoogleIntelligence index45.4→46.7+2.9%
ScoreGemini 3.5 Flash (minimal)GoogleIntelligence index34.9→35.8+2.6%
ScoreGemini 3.5 Flash-LiteGoogleIntelligence index36.5→37.4+2.5%
ScoreGemini 3.5 Flash-LiteGoogleAgentic index26.8→27.2+1.5%
ScoreGemini 3.6 Flash (high)GoogleIntelligence index50.1→51.6+3.0%
ScoreGemini 3.6 Flash (high)GoogleAgentic index38.7→40.5+4.7%
ScoreGemma 3 270MGoogleIntelligence index2.4→2-17%
ScoreGemma 3 4B InstructGoogleIntelligence index1.1→1-9.1%
ScoreGemma 3n E4B InstructGoogleIntelligence index1.2→1-17%
ScoreGemma 3n E4B Instruct Preview (May '25)GoogleIntelligence index4.6→4.2-8.7%
ScoreGemma 4 12B (Reasoning)GoogleIntelligence index21.8→22.2+1.8%
ScoreGemma 4 12B (Reasoning)GoogleAgentic index7.9→Metric withdrawn
ScoreGemma 4 26B A4B (Non-reasoning)GoogleIntelligence index20.1→20.4+1.5%
ScoreGemma 4 26B A4B (Reasoning)GoogleIntelligence index25.7→26.1+1.6%
ScoreGemma 4 31B (Non-reasoning)GoogleIntelligence index21.8→22.3+2.3%
ScoreGemma 4 31B (Non-reasoning)GoogleAgentic index11→11.1+0.9%
ScoreGemma 4 31B (Reasoning)GoogleIntelligence index29.4→30+2.0%
ScoreGemma 4 31B (Reasoning)GoogleAgentic index14.4→Metric withdrawn
ScoreGemma 4 E2B (Non-reasoning)GoogleIntelligence index6.4→6.2-3.1%
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index9.7→9.5-2.1%
ScoreGemma 4 E2B (Reasoning)GoogleAgentic index1.5→Metric withdrawn
ScoreGemma 4 E4B (Non-reasoning)GoogleIntelligence index8.9→8.7-2.2%
ScoreGemma 4 E4B (Reasoning)GoogleAgentic index1.8→Metric withdrawn
ScoreGLM 5V Turbo (Reasoning)Z AIIntelligence index34.5→35.3+2.3%
ScoreGLM-4.5 (Reasoning)Z AIIntelligence index19.5→19.7+1.0%
ScoreGLM-4.5-AirZ AIIntelligence index16.5→16.7+1.2%
ScoreGLM-4.5V (Non-reasoning)Z AIIntelligence index7→6.8-2.9%
ScoreGLM-4.5V (Reasoning)Z AIIntelligence index9.1→9-1.1%
ScoreGLM-4.6 (Non-reasoning)Z AIIntelligence index23→23.4+1.7%
ScoreGLM-4.6 (Reasoning)Z AIIntelligence index28.7→29.3+2.1%
ScoreGLM-4.6 (Reasoning)Z AIAgentic index17.7→18.6+5.1%
ScoreGLM-4.6V (Non-reasoning)Z AIIntelligence index11→10.9-0.9%
ScoreGLM-4.6V (Reasoning)Z AIIntelligence index16.8→16.9+0.6%
ScoreGLM-4.7 (Non-reasoning)Z AIIntelligence index26.6→27.1+1.9%
ScoreGLM-4.7 (Reasoning)Z AIIntelligence index33.7→34.5+2.4%
ScoreGLM-4.7 (Reasoning)Z AIAgentic index25.4→26.2+3.1%
ScoreGLM-4.7-Flash (Non-reasoning)Z AIIntelligence index15.5→15.6+0.6%
ScoreGLM-4.7-Flash (Reasoning)Z AIIntelligence index22.9→23.3+1.7%
ScoreGLM-5 (Non-reasoning)Z AIIntelligence index32.4→33.2+2.5%
ScoreGLM-5 (Reasoning)Z AIIntelligence index39.5→40.6+2.8%
ScoreGLM-5-TurboZ AIIntelligence index38.1→39.1+2.6%
ScoreGLM-5.1 (Non-reasoning)Z AIIntelligence index35.4→36.3+2.5%
ScoreGLM-5.1 (Reasoning)Z AIIntelligence index40.2→41+2.0%
ScoreGLM-5.1 (Reasoning)Z AIAgentic index29.9→30.6+2.3%
ScoreGLM-5.2 (max)Z AIIntelligence index51.1→52.6+2.9%
ScoreGLM-5.2 (max)Z AIAgentic index43.1→45.7+6.0%
ScoreGLM-5.2 (Non-reasoning)Z AIIntelligence index34.1→34.8+2.1%
ScoreGLM-5.2 (Non-reasoning)Z AIAgentic index34.8→35.1+0.9%
ScoreGPT-3.5 TurboOpenAIIntelligence index3.6→3.2-11%
ScoreGPT-4OpenAIIntelligence index7→6.8-2.9%
ScoreGPT-4 TurboOpenAIIntelligence index7.9→7.7-2.5%
ScoreGPT-4.1OpenAIIntelligence index19.4→19.6+1.0%
ScoreGPT-4.1 miniOpenAIAgentic index1.7→1.8+5.9%
ScoreGPT-4o (Aug '24)OpenAIIntelligence index9.6→9.4-2.1%
ScoreGPT-4o (ChatGPT)OpenAIIntelligence index8.2→8.1-1.2%
ScoreGPT-4o (May '24)OpenAIIntelligence index8.6→8.4-2.3%
ScoreGPT-4o (Nov '24)OpenAIIntelligence index11.2→11.1-0.9%
ScoreGPT-4o miniOpenAIIntelligence index6.9→6.7-2.9%
ScoreGPT-5 (ChatGPT)OpenAIIntelligence index15.3→15.4+0.7%
ScoreGPT-5 (high)OpenAIIntelligence index34.7→35.3+1.7%
ScoreGPT-5 (high)OpenAIAgentic index25.7→26.5+3.1%
ScoreGPT-5 (low)OpenAIIntelligence index31.2→31.9+2.2%
ScoreGPT-5 (medium)OpenAIIntelligence index33.7→34.6+2.7%
ScoreGPT-5 (minimal)OpenAIIntelligence index17.2→17.3+0.6%
ScoreGPT-5 Codex (high)OpenAIIntelligence index36.1→37+2.5%
ScoreGPT-5 mini (high)OpenAIIntelligence index25.3→25.8+2.0%
ScoreGPT-5 mini (high)OpenAIAgentic index19.4→19.6+1.0%
ScoreGPT-5 mini (medium)OpenAIIntelligence index30.9→31.6+2.3%
ScoreGPT-5 nano (high)OpenAIIntelligence index19.9→20.1+1.0%
ScoreGPT-5 nano (medium)OpenAIIntelligence index19→19.2+1.1%
ScoreGPT-5 nano (minimal)OpenAIIntelligence index8→7.8-2.5%
ScoreGPT-5.1 (high)OpenAIIntelligence index36.9→37.5+1.6%
ScoreGPT-5.1 (high)OpenAIAgentic index21→21.6+2.9%
ScoreGPT-5.1 (Non-reasoning)OpenAIIntelligence index20.4→20.7+1.5%
ScoreGPT-5.1 Codex (high)OpenAIIntelligence index34.7→35.6+2.6%
ScoreGPT-5.1 Codex mini (high)OpenAIIntelligence index30.6→31.3+2.3%
ScoreGPT-5.2 (medium)OpenAIIntelligence index38→38.9+2.4%
ScoreGPT-5.2 (Non-reasoning)OpenAIIntelligence index26→26.5+1.9%
ScoreGPT-5.2 (xhigh)OpenAIIntelligence index42.2→43.3+2.6%
ScoreGPT-5.2 Codex (xhigh)OpenAIIntelligence index40.1→41.2+2.7%
ScoreGPT-5.3 Codex (xhigh)OpenAIIntelligence index44.3→45.5+2.7%
ScoreGPT-5.4 (low)OpenAIIntelligence index39.1→40.2+2.8%
ScoreGPT-5.4 (Non-reasoning)OpenAIIntelligence index27.7→28.3+2.2%
ScoreGPT-5.4 (xhigh)OpenAIIntelligence index51.4→53.1+3.3%
ScoreGPT-5.4 (xhigh)OpenAIAgentic index41.1→44.2+7.5%
ScoreGPT-5.4 mini (medium)OpenAIIntelligence index29.8→30.5+2.3%
ScoreGPT-5.4 mini (Non-Reasoning)OpenAIIntelligence index16.6→16.8+1.2%
ScoreGPT-5.4 mini (xhigh)OpenAIIntelligence index40→40.9+2.2%
ScoreGPT-5.4 mini (xhigh)OpenAIAgentic index30.2→31.5+4.3%
ScoreGPT-5.4 nano (medium)OpenAIIntelligence index30.2→30.8+2.0%
ScoreGPT-5.4 nano (Non-Reasoning)OpenAIIntelligence index17.6→17.8+1.1%
ScoreGPT-5.4 nano (xhigh)OpenAIIntelligence index38.2→39.7+3.9%
ScoreGPT-5.4 nano (xhigh)OpenAIAgentic index27.5→29.7+8.0%
ScoreGPT-5.5 (high)OpenAIIntelligence index53.1→54.7+3.0%
ScoreGPT-5.5 (high)OpenAIAgentic index43.5→45.9+5.5%
ScoreGPT-5.5 (low)OpenAIIntelligence index43.5→44.5+2.3%
ScoreGPT-5.5 (low)OpenAIAgentic index30.4→31.7+4.3%
ScoreGPT-5.5 (medium)OpenAIIntelligence index50.4→51.4+2.0%
ScoreGPT-5.5 (medium)OpenAIAgentic index37.8→39.2+3.7%
ScoreGPT-5.5 (Non-reasoning)OpenAIIntelligence index35.4→35.8+1.1%
ScoreGPT-5.5 (Non-reasoning)OpenAIAgentic index25.8→26.1+1.2%
ScoreGPT-5.5 (xhigh)OpenAIIntelligence index54.8→56.3+2.7%
ScoreGPT-5.5 (xhigh)OpenAIAgentic index44.9→47.4+5.6%
ScoreGPT-5.5 Instant (June 2026)OpenAIIntelligence index28.9→29.2+1.0%
ScoreGPT-5.5 Instant (June 2026)OpenAIAgentic index11→11.1+0.9%
ScoreGPT-5.5 Instant (May 2026)OpenAIIntelligence index33.5→34.3+2.4%
ScoreGPT-5.6 Luna (high)OpenAIIntelligence index46.1→47+2.0%
ScoreGPT-5.6 Luna (high)OpenAIAgentic index40.1→41+2.2%
ScoreGPT-5.6 Luna (low)OpenAIIntelligence index33.3→33.9+1.8%
ScoreGPT-5.6 Luna (low)OpenAIAgentic index25.4→25.7+1.2%
ScoreGPT-5.6 Luna (max)OpenAIIntelligence index51.2→52.3+2.1%
ScoreGPT-5.6 Luna (max)OpenAIAgentic index45.6→46.9+2.9%
ScoreGPT-5.6 Luna (medium)OpenAIIntelligence index38.1→38.9+2.1%
ScoreGPT-5.6 Luna (medium)OpenAIAgentic index31→31.8+2.6%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIIntelligence index26.6→26.8+0.8%
ScoreGPT-5.6 Luna (Non-reasoning)OpenAIAgentic index22→22.2+0.9%
ScoreGPT-5.6 Luna (xhigh)OpenAIIntelligence index49.1→50.1+2.0%
ScoreGPT-5.6 Luna (xhigh)OpenAIAgentic index42.9→44.4+3.5%
ScoreGPT-5.6 Sol (high)OpenAIIntelligence index55.9→57.3+2.5%
ScoreGPT-5.6 Sol (high)OpenAIAgentic index48.5→50.6+4.3%
ScoreGPT-5.6 Sol (low)OpenAIIntelligence index49.4→50.7+2.6%
ScoreGPT-5.6 Sol (low)OpenAIAgentic index40→41.6+4.0%
ScoreGPT-5.6 Sol (max)OpenAIIntelligence index58.9→60.9+3.4%
ScoreGPT-5.6 Sol (max)OpenAIAgentic index54→57.8+7.0%
ScoreGPT-5.6 Sol (medium)OpenAIIntelligence index53.6→55.6+3.7%
ScoreGPT-5.6 Sol (medium)OpenAIAgentic index44.5→47.9+7.6%
ScoreGPT-5.6 Sol (Non-reasoning)OpenAIIntelligence index41.2→41.9+1.7%
ScoreGPT-5.6 Sol (Non-reasoning)OpenAIAgentic index34.9→36+3.2%
ScoreGPT-5.6 Sol (xhigh)OpenAIIntelligence index57.7→59+2.3%
ScoreGPT-5.6 Sol (xhigh)OpenAIAgentic index51.8→53.6+3.5%
ScoreGPT-5.6 Terra (high)OpenAIIntelligence index49→50.1+2.2%
ScoreGPT-5.6 Terra (high)OpenAIAgentic index41.3→43.5+5.3%
ScoreGPT-5.6 Terra (low)OpenAIIntelligence index40.5→41.3+2.0%
ScoreGPT-5.6 Terra (low)OpenAIAgentic index30.6→31.5+2.9%
ScoreGPT-5.6 Terra (max)OpenAIIntelligence index55→56.6+2.9%
ScoreGPT-5.6 Terra (max)OpenAIAgentic index47.4→50.2+5.9%
ScoreGPT-5.6 Terra (medium)OpenAIIntelligence index45.6→46.8+2.6%
ScoreGPT-5.6 Terra (medium)OpenAIAgentic index37→39.1+5.7%
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIIntelligence index34→34.6+1.8%
ScoreGPT-5.6 Terra (Non-reasoning)OpenAIAgentic index29.3→30.1+2.7%
ScoreGPT-5.6 Terra (xhigh)OpenAIIntelligence index51.6→52.8+2.3%
ScoreGPT-5.6 Terra (xhigh)OpenAIAgentic index44.7→46.5+4.0%
Scoregpt-oss-120b (high)OpenAIIntelligence index23.8→24.1+1.3%
Scoregpt-oss-120b (high)OpenAIAgentic index13.2→13.4+1.5%
Scoregpt-oss-20b (high)OpenAIIntelligence index14.9→15.2+2.0%
Scoregpt-oss-20b (low)OpenAIIntelligence index14.3→14.4+0.7%
ScoreGranite 3.3 8B (Non-reasoning)IBMIntelligence index1.8→1.3-28%
ScoreGranite 4.0 1BIBMIntelligence index2.1→1.6-24%
ScoreGranite 4.0 H 1BIBMIntelligence index2.7→2.2-19%
ScoreGranite 4.0 H SmallIBMIntelligence index5.2→4.9-5.8%
ScoreGranite 4.0 MicroIBMIntelligence index2.4→2-17%
ScoreGranite 4.1 30BIBMIntelligence index8.9→8.7-2.2%
ScoreGranite 4.1 30BIBMAgentic index1.3→Metric withdrawn
ScoreGranite 4.1 3BIBMIntelligence index4.7→4.4-6.4%
ScoreGranite 4.1 3BIBMAgentic index0.4→Metric withdrawn
ScoreGranite 4.1 8BIBMIntelligence index6.7→6.4-4.5%
ScoreGrok 2 (Dec '24)SpaceXAIIntelligence index8→7.8-2.5%
ScoreGrok 3SpaceXAIIntelligence index18.4→18.6+1.1%
ScoreGrok 3 mini Reasoning (high)SpaceXAIIntelligence index22.5→22.9+1.8%
ScoreGrok 3 Reasoning BetaSpaceXAIIntelligence index15.1→15.2+0.7%
ScoreGrok 4SpaceXAIIntelligence index33.3→34.1+2.4%
ScoreGrok 4 Fast (Non-reasoning)SpaceXAIIntelligence index16.5→16.6+0.6%
ScoreGrok 4 Fast (Reasoning)SpaceXAIIntelligence index27.4→27.9+1.8%
ScoreGrok 4.1 Fast (Non-reasoning)SpaceXAIIntelligence index16.9→17+0.6%
ScoreGrok 4.1 Fast (Reasoning)SpaceXAIIntelligence index30.6→31.3+2.3%
ScoreGrok 4.20 0309 (Non-reasoning)SpaceXAIIntelligence index22.5→22.9+1.8%
ScoreGrok 4.20 0309 (Reasoning)SpaceXAIIntelligence index36.5→37.4+2.5%
ScoreGrok 4.20 0309 v2 (Non-reasoning)SpaceXAIIntelligence index21.8→22.2+1.8%
ScoreGrok 4.20 0309 v2 (Reasoning)SpaceXAIIntelligence index37→38+2.7%
ScoreGrok 4.3 (high)SpaceXAIIntelligence index37.6→37.9+0.8%
ScoreGrok 4.3 (high)SpaceXAIAgentic index24.1→24.2+0.4%
ScoreGrok 4.3 (low)SpaceXAIIntelligence index35.4→36.3+2.5%
ScoreGrok 4.3 (medium)SpaceXAIIntelligence index36→36.9+2.5%
ScoreGrok 4.3 (Non-reasoning)SpaceXAIIntelligence index24.8→25+0.8%
ScoreGrok 4.3 (Non-reasoning)SpaceXAIAgentic index22.8→23+0.9%
ScoreGrok 4.5 (high)SpaceXAIIntelligence index53.8→55.8+3.7%
ScoreGrok 4.5 (high)SpaceXAIAgentic index45.7→48.9+7.0%
ScoreGrok BetaSpaceXAIIntelligence index7.5→7.3-2.7%
ScoreGrok Build 0.1 0616SpaceXAIIntelligence index39.8→40.7+2.3%
ScoreGrok Build 0.1 0616SpaceXAIAgentic index28→28.9+3.2%
ScoreGrok Code Fast 1SpaceXAIIntelligence index21.6→22+1.9%
ScoreGrok-1SpaceXAIIntelligence index6→5.8-3.3%
ScoreHermes 3 - Llama-3.1 70BNous ResearchIntelligence index5.1→4.8-5.9%
ScoreHermes 4 - Llama-3.1 405B (Non-reasoning)Nous ResearchIntelligence index8.8→8.6-2.3%
ScoreHermes 4 - Llama-3.1 405B (Reasoning)Nous ResearchIntelligence index9→8.8-2.2%
ScoreHermes 4 - Llama-3.1 70B (Non-reasoning)Nous ResearchIntelligence index6.9→6.7-2.9%
ScoreHermes 4 - Llama-3.1 70B (Reasoning)Nous ResearchIntelligence index10→9.9-1.0%
ScoreHy3TencentIntelligence index41.2→42.2+2.4%
ScoreHy3TencentAgentic index30.7→31.4+2.3%
ScoreHy3-preview (Non-reasoning)TencentIntelligence index26.1→26.6+1.9%
ScoreHy3-preview (Reasoning)TencentIntelligence index33.6→34.4+2.4%
ScoreHyperCLOVA X SEED Think (32B)NaverIntelligence index17→17.2+1.2%
ScoreHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingIntelligence index17.8→18.3+2.8%
ScoreHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse ComputingAgentic index6.7→6.8+1.5%
ScoreInkling (xhigh)Thinking MachinesIntelligence index40.7→42.3+3.9%
ScoreInkling (xhigh)Thinking MachinesAgentic index32.3→34.1+5.6%
ScoreInkling SmallThinking MachinesIntelligence index40.2→41.2+2.5%
ScoreInkling SmallThinking MachinesAgentic index30.8→31.9+3.6%
ScoreINTELLECT-3 (based on GLM-4.5-Air)Prime IntellectIntelligence index15.6→15.7+0.6%
ScoreJamba 1.5 LargeAI21 LabsIntelligence index5.1→4.8-5.9%
ScoreJamba 1.5 MiniAI21 LabsIntelligence index2.7→2.3-15%
ScoreJamba 1.6 LargeAI21 LabsIntelligence index5→4.7-6.0%
ScoreJamba 1.6 MiniAI21 LabsIntelligence index2.6→2.1-19%
ScoreJamba 1.7 LargeAI21 LabsIntelligence index5.3→5-5.7%
ScoreJamba 1.7 MiniAI21 LabsIntelligence index2.7→2.3-15%
ScoreJamba Reasoning 3BAI21 LabsIntelligence index4.1→3.8-7.3%
ScoreJT-35B-FlashChina MobileIntelligence index28.4→29+2.1%
ScoreJT-4.1 Flash 236B A21BChina MobileIntelligence index38.8→39.9+2.8%
ScoreJT-4.1 Flash 236B A21BChina MobileAgentic index34.9→Metric withdrawn
ScoreJT-MINIChina MobileIntelligence index18.5→18.8+1.6%
ScoreK-EXAONE (Non-reasoning)LG AI ResearchIntelligence index16.7→16.9+1.2%
ScoreK-EXAONE (Reasoning)LG AI ResearchIntelligence index22.1→22.5+1.8%
ScoreK-EXAONE (Reasoning)LG AI ResearchAgentic index8→Metric withdrawn
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsIntelligence index17.3→17.4+0.6%
ScoreK2 Think V2MBZUAI Institute of Foundation ModelsAgentic index1.8→Metric withdrawn
ScoreK2-V2 (low)MBZUAI Institute of Foundation ModelsIntelligence index8.6→8.4-2.3%
ScoreKAT Coder Pro V2KwaiKATIntelligence index33.7→34.5+2.4%
ScoreKAT Coder Pro V2KwaiKATAgentic index15.5→Metric withdrawn
ScoreKAT-Coder-Pro V1KwaiKATIntelligence index28.3→28.9+2.1%
ScoreKimi K2KimiIntelligence index19.4→19.7+1.5%
ScoreKimi K2 0905KimiIntelligence index23.5→24+2.1%
ScoreKimi K2 ThinkingKimiIntelligence index32.7→33.5+2.4%
ScoreKimi K2.5 (Non-reasoning)KimiIntelligence index29.4→30.1+2.4%
ScoreKimi K2.5 (Reasoning)KimiIntelligence index35.4→36+1.7%
ScoreKimi K2.6KimiIntelligence index44.2→45.1+2.0%
ScoreKimi K2.6KimiAgentic index30.3→31.2+3.0%
ScoreKimi K2.6 (Non-reasoning)KimiIntelligence index34.6→35.4+2.3%
ScoreKimi K2.7 CodeKimiIntelligence index41.9→43+2.6%
ScoreKimi K2.7 CodeKimiAgentic index29.6→30.3+2.4%
ScoreKimi K3 (low)KimiIntelligence index46.6→48.3+3.6%
ScoreKimi K3 (low)KimiAgentic index37→39.6+7.0%
ScoreKimi K3 (max)KimiIntelligence index57.1→59.7+4.6%
ScoreKimi K3 (max)KimiAgentic index50.1→54.3+8.4%
ScoreKimi Linear 48B A3B InstructKimiIntelligence index8.5→8.4-1.2%
ScoreLFM 40BLiquid AIIntelligence index3.4→3-12%
ScoreLFM2 1.2BLiquid AIIntelligence index1.1→1-9.1%
ScoreLFM2 2.6BLiquid AIIntelligence index2.7→2.3-15%
ScoreLFM2 24B A2BLiquid AIIntelligence index4.9→4.6-6.1%
ScoreLFM2 8B A1BLiquid AIIntelligence index1.8→1.3-28%
ScoreLFM2.5-1.2B-InstructLiquid AIIntelligence index2.7→2.3-15%
ScoreLFM2.5-1.2B-ThinkingLiquid AIIntelligence index2.7→2.3-15%
ScoreLFM2.5-8B-A1BLiquid AIIntelligence index8.3→8.1-2.4%
ScoreLing 2.6 FlashInclusionAIIntelligence index14.1→14.2+0.7%
ScoreLing-1TInclusionAIIntelligence index12.8→12.7-0.8%
ScoreLing-2.6-1TInclusionAIIntelligence index26.1→26.6+1.9%
ScoreLing-flash-2.0InclusionAIIntelligence index9.7→9.6-1.0%
ScoreLing-mini-2.0InclusionAIIntelligence index3.8→3.4-11%
ScoreLlama 2 Chat 13BMetaIntelligence index3→2.6-13%
ScoreLlama 2 Chat 70BMetaIntelligence index3→2.6-13%
ScoreLlama 2 Chat 7BMetaIntelligence index4.3→3.9-9.3%
ScoreLlama 3 Instruct 70BMetaIntelligence index3.5→3.1-11%
ScoreLlama 3 Instruct 8BMetaIntelligence index1.2→1-17%
ScoreLlama 3.1 Instruct 405BMetaIntelligence index8.5→8.3-2.4%
ScoreLlama 3.1 Instruct 70BMetaIntelligence index6.8→6.5-4.4%
ScoreLlama 3.1 Instruct 8BMetaIntelligence index7.6→7.4-2.6%
ScoreLlama 3.1 Instruct 8BMetaAgentic index0.5→Metric withdrawn
ScoreLlama 3.1 Nemotron Instruct 70BNVIDIAIntelligence index7.6→7.4-2.6%
ScoreLlama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIAIntelligence index8.5→8.4-1.2%
ScoreLlama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIAIntelligence index9.1→8.9-2.2%
ScoreLlama 3.1 Tulu3 405BAllen Institute for AIIntelligence index8.3→8.1-2.4%
ScoreLlama 3.2 Instruct 11B (Vision)MetaIntelligence index3.3→3-9.1%
ScoreLlama 3.2 Instruct 1BMetaIntelligence index1.1→1-9.1%
ScoreLlama 3.2 Instruct 3BMetaIntelligence index4.2→3.9-7.1%
ScoreLlama 3.2 Instruct 90B (Vision)MetaIntelligence index6.2→6-3.2%
ScoreLlama 3.3 Instruct 70BMetaIntelligence index9.4→9.3-1.1%
ScoreLlama 3.3 Instruct 70BMetaAgentic index0.3→Metric withdrawn
ScoreLlama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIAIntelligence index8.5→8.3-2.4%
ScoreLlama 4 MaverickMetaIntelligence index14.3→14.5+1.4%
ScoreLlama 4 MaverickMetaAgentic index1.3→1.2-7.7%
ScoreLlama 4 ScoutMetaIntelligence index10→10.3+3.0%
ScoreLlama 65BMetaIntelligence index2.1→1.7-19%
ScoreLlama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIAIntelligence index8.7→8.5-2.3%
ScoreLongCat 2.0LongCatIntelligence index33.5→34.3+2.4%
ScoreLongCat 2.0LongCatAgentic index21.8→Metric withdrawn
ScoreLongCat Flash LiteLongCatIntelligence index17.2→17.4+1.2%
ScoreMagistral Medium 1.2MistralIntelligence index17.9→18+0.6%
ScoreMagistral Small 1MistralIntelligence index10.7→10.6-0.9%
ScoreMagistral Small 1.2MistralIntelligence index11.3→11.5+1.8%
ScoreMagistral Small 1.2MistralAgentic index1.4→1.5+7.1%
ScoreMercury 2InceptionIntelligence index21.4→21.9+2.3%
ScoreMercury 2InceptionAgentic index9.6→9.5-1.0%
ScoreMi:dm K 2.5 ProKorea TelecomIntelligence index16.4→16.6+1.2%
ScoreMiMo-V2-Flash (Feb 2026)XiaomiIntelligence index33.2→34+2.4%
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiIntelligence index24.7→25.1+1.6%
ScoreMiMo-V2-Flash (Non-reasoning)XiaomiAgentic index12→Metric withdrawn
ScoreMiMo-V2-Flash (Reasoning)XiaomiIntelligence index31.2→31.9+2.2%
ScoreMiMo-V2-OmniXiaomiIntelligence index35→35.9+2.6%
ScoreMiMo-V2-Omni-0327XiaomiIntelligence index36.4→37.3+2.5%
ScoreMiMo-V2-ProXiaomiIntelligence index40.3→41.4+2.7%
ScoreMiMo-V2.5XiaomiIntelligence index37.2→38+2.2%
ScoreMiMo-V2.5XiaomiAgentic index23.7→24.4+3.0%
ScoreMiMo-V2.5-ProXiaomiIntelligence index42.2→42.9+1.7%
ScoreMiMo-V2.5-ProXiaomiAgentic index29.1→29.5+1.4%
ScoreMiMo-V2.5-Pro (Non-reasoning)XiaomiIntelligence index27.9→28.4+1.8%
ScoreMiniCPM-V 4.6 1.3BOpenBMBIntelligence index4.2→3.8-9.5%
ScoreMiniCPM-V 4.6 1.3BOpenBMBAgentic index1.4→Metric withdrawn
ScoreMiniCPM5-1B (Reasoning)OpenBMBIntelligence index12→11.9-0.8%
ScoreMiniMax M1 40kMiniMaxIntelligence index14.4→14.5+0.7%
ScoreMiniMax M1 80kMiniMaxIntelligence index17.7→17.9+1.1%
ScoreMiniMax-M2MiniMaxIntelligence index28.3→28.9+2.1%
ScoreMiniMax-M2.1MiniMaxIntelligence index31.4→32.1+2.2%
ScoreMiniMax-M2.5MiniMaxIntelligence index33.7→34.5+2.4%
ScoreMiniMax-M2.7MiniMaxIntelligence index38.1→38.9+2.1%
ScoreMiniMax-M2.7MiniMaxAgentic index25.6→25.9+1.2%
ScoreMiniMax-M3MiniMaxIntelligence index44.4→45.4+2.3%
ScoreMiniMax-M3MiniMaxAgentic index35.4→36.1+2.0%
ScoreMinistral 3 14BMistralIntelligence index11.1→11.2+0.9%
ScoreMinistral 3 3BMistralIntelligence index6.8→7.1+4.4%
ScoreMistral 7B InstructMistralIntelligence index2.1→1.7-19%
ScoreMistral Large (Feb '24)MistralIntelligence index4.4→4.1-6.8%
ScoreMistral Large 2 (Jul '24)MistralIntelligence index7.3→7-4.1%
ScoreMistral Large 2 (Nov '24)MistralIntelligence index9.1→9-1.1%
ScoreMistral MediumMistralIntelligence index3.6→3.2-11%
ScoreMistral Medium 3.1MistralAgentic index6.2→6.1-1.6%
ScoreMistral Medium 3.5MistralIntelligence index29.9→30.4+1.7%
ScoreMistral Medium 3.5MistralAgentic index19→19.2+1.1%
ScoreMistral SabaMistralIntelligence index6.4→6.2-3.1%
ScoreMistral Small (Feb '24)MistralIntelligence index3.6→3.2-11%
ScoreMistral Small (Sep '24)MistralIntelligence index4.7→4.3-8.5%
ScoreMistral Small 3MistralIntelligence index6.9→6.7-2.9%
ScoreMistral Small 3.1MistralIntelligence index14.7→14.9+1.4%
ScoreMistral Small 3.1MistralAgentic index5.2→5.3+1.9%
ScoreMistral Small 3.2MistralIntelligence index10.6→10.7+0.9%
ScoreMistral Small 3.2MistralAgentic index2→2.1+5.0%
ScoreMistral Small 4 (Non-reasoning)MistralIntelligence index12.4→12.3-0.8%
ScoreMistral Small 4 (Reasoning)MistralIntelligence index19.6→19.7+0.5%
ScoreMistral Small 4 (Reasoning)MistralAgentic index4.7→4.6-2.1%
ScoreMixtral 8x22B InstructMistralIntelligence index4.4→4-9.1%
ScoreMixtral 8x7B InstructMistralIntelligence index2.4→2-17%
ScoreMolmo 7B-DAllen Institute for AIIntelligence index3.8→3.4-11%
ScoreMolmo2-8BAllen Institute for AIIntelligence index2→1.6-20%
ScoreMotif 3 (Beta)Motif TechnologiesIntelligence index44.1→45.3+2.7%
ScoreMotif 3 (Beta)Motif TechnologiesAgentic index34.9→Metric withdrawn
ScoreMuse SparkMetaIntelligence index43.1→44.3+2.8%
ScoreMuse SparkMetaAgentic index28.7→Metric withdrawn
ScoreMuse Spark 1.1 (xhigh)MetaIntelligence index50.6→53.2+5.1%
ScoreMuse Spark 1.1 (xhigh)MetaAgentic index37.5→39.7+5.9%
ScoreNanbeige4.1-3BNanbeigeIntelligence index11.1→11-0.9%
ScoreNanbeige4.1-3BNanbeigeAgentic index0→Metric withdrawn
ScoreNemotron 3 Nano Omni 30B A3B ReasoningNVIDIAIntelligence index14.9→15+0.7%
ScoreNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAIntelligence index37.8→38.3+1.3%
ScoreNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAAgentic index27.4→27.5+0.4%
ScoreNemotron Cascade 2 30B A3BNVIDIAIntelligence index17.6→17.8+1.1%
ScoreNemotron Cascade 2 30B A3BNVIDIAAgentic index3.7→Metric withdrawn
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIIntelligence index41→42.1+2.7%
ScoreNex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGIAgentic index31→Metric withdrawn
ScoreNorth Mini CodeCohereIntelligence index19.8→20.2+2.0%
ScoreNova 2.0 Lite (high)AmazonIntelligence index18.2→20.8+14%
ScoreNova 2.0 Lite (high)AmazonAgentic index3.1→Metric withdrawn
ScoreNova 2.0 Lite (low)AmazonIntelligence index17.8→18+1.1%
ScoreNova 2.0 Lite (medium)AmazonIntelligence index19→19.2+1.1%
ScoreNova 2.0 Omni (low)AmazonIntelligence index16.6→16.7+0.6%
ScoreNova 2.0 Omni (medium)AmazonIntelligence index20.9→21.3+1.9%
ScoreNova 2.0 Omni (Non-reasoning)AmazonIntelligence index10.5→10.4-1.0%
ScoreNova 2.0 Pro Preview (low)AmazonIntelligence index19.6→19.8+1.0%
ScoreNova 2.0 Pro Preview (low)AmazonAgentic index6.4→Metric withdrawn
ScoreNova 2.0 Pro Preview (medium)AmazonIntelligence index21.8→22.1+1.4%
ScoreNova 2.0 Pro Preview (medium)AmazonAgentic index7→Metric withdrawn
ScoreNova 2.0 Pro Preview (Non-reasoning)AmazonAgentic index2.9→Metric withdrawn
ScoreNova LiteAmazonIntelligence index6.9→6.7-2.9%
ScoreNova MicroAmazonIntelligence index4.7→4.4-6.4%
ScoreNova ProAmazonIntelligence index7.7→7.5-2.6%
ScoreNVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIAIntelligence index7.4→7.2-2.7%
ScoreNVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIAIntelligence index14.2→14.5+2.1%
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAIntelligence index8.7→8.6-1.1%
ScoreNVIDIA Nemotron 3 Nano 4BNVIDIAAgentic index2.3→Metric withdrawn
ScoreNemotron 3 Super 120B A12B (Reasoning)NVIDIAIntelligence index25.4→25.7+1.2%
ScoreNemotron 3 Super 120B A12B (Reasoning)NVIDIAAgentic index8.7→8.8+1.1%
ScoreNVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIAIntelligence index4.6→4.2-8.7%
ScoreNVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIAIntelligence index9→8.8-2.2%
ScoreNVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIAIntelligence index7.4→7.2-2.7%
ScoreNVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIAIntelligence index8.8→8.7-1.1%
Scoreo1OpenAIIntelligence index23.4→23.9+2.1%
Scoreo1-previewOpenAIIntelligence index17→17.2+1.2%
Scoreo1-proOpenAIIntelligence index18.9→19.1+1.1%
Scoreo3OpenAIIntelligence index30.4→31.1+2.3%
Scoreo3-miniOpenAIIntelligence index19→19.2+1.1%
Scoreo3-mini (high)OpenAIIntelligence index15.6→15.7+0.6%
Scoreo3-proOpenAIIntelligence index32.5→33.3+2.5%
Scoreo4-mini (high)OpenAIIntelligence index25.6→26.1+2.0%
ScoreOLMo 2 32BAllen Institute for AIIntelligence index5→4.7-6.0%
ScoreOLMo 2 7BAllen Institute for AIIntelligence index3.9→3.5-10%
ScoreOlmo 3 32B ThinkAllen Institute for AIIntelligence index6.4→6.1-4.7%
ScoreOlmo 3 7B InstructAllen Institute for AIIntelligence index2.8→2.4-14%
ScoreOlmo 3 7B ThinkAllen Institute for AIIntelligence index4→3.6-10.0%
ScoreOlmo 3.1 32B InstructAllen Institute for AIIntelligence index6.5→6.2-4.6%
ScoreOlmo 3.1 32B ThinkAllen Institute for AIIntelligence index8.1→7.9-2.5%
ScoreOpenChat 3.5 (1210)OpenChatIntelligence index3→2.6-13%
ScorePALM-2GoogleIntelligence index3.2→2.8-13%
ScorePhi-3 Mini Instruct 3.8BMicrosoftIntelligence index4.6→4.3-6.5%
ScorePhi-4MicrosoftIntelligence index4.9→4.6-6.1%
ScorePhi-4 Mini InstructMicrosoftIntelligence index6→5.7-5.0%
ScorePhi-4 Mini InstructMicrosoftAgentic index0.3→Metric withdrawn
ScorePhi-4 Multimodal InstructMicrosoftIntelligence index4.5→4.2-6.7%
ScorePixtral LargeMistralIntelligence index8.1→8-1.2%
ScoreQwen Chat 14BAlibabaIntelligence index2.1→1.7-19%
ScoreQwen Chat 72BAlibabaIntelligence index3.4→3-12%
ScoreQwen1.5 Chat 110BAlibabaIntelligence index4.1→3.7-9.8%
ScoreQwen2 Instruct 72BAlibabaIntelligence index6→5.7-5.0%
ScoreQwen2.5 Coder Instruct 32BAlibabaIntelligence index7.1→6.9-2.8%
ScoreQwen2.5 Coder Instruct 7BAlibabaIntelligence index4.5→4.1-8.9%
ScoreQwen2.5 Instruct 32BAlibabaIntelligence index7.5→7.2-4.0%
ScoreQwen2.5 Instruct 72BAlibabaIntelligence index9.6→9.4-2.1%
ScoreQwen2.5 MaxAlibabaIntelligence index10.2→10.1-1.0%
ScoreQwen2.5 TurboAlibabaIntelligence index6.3→6-4.8%
ScoreQwen3 0.6B (Reasoning)AlibabaIntelligence index1.3→1-23%
ScoreQwen3 1.7B (Non-reasoning)AlibabaIntelligence index1.5→1.1-27%
ScoreQwen3 1.7B (Reasoning)AlibabaIntelligence index2.6→2.2-15%
ScoreQwen3 14B (Non-reasoning)AlibabaIntelligence index7→6.8-2.9%
ScoreQwen3 14B (Reasoning)AlibabaAgentic index1.8→1.9+5.6%
ScoreQwen3 235B A22B (Non-reasoning)AlibabaIntelligence index10.9→10.8-0.9%
ScoreQwen3 235B A22B (Reasoning)AlibabaIntelligence index13.4→13.5+0.7%
ScoreQwen3 235B A22B 2507 (Reasoning)AlibabaIntelligence index19.6→19.9+1.5%
ScoreQwen3 235B A22B 2507 InstructAlibabaIntelligence index18.2→18.4+1.1%
ScoreQwen3 30B A3B (Non-reasoning)AlibabaIntelligence index6.8→6.6-2.9%
ScoreQwen3 30B A3B (Reasoning)AlibabaIntelligence index9.3→9.2-1.1%
ScoreQwen3 30B A3B 2507 (Reasoning)AlibabaIntelligence index14.4→14.6+1.4%
ScoreQwen3 30B A3B 2507 InstructAlibabaIntelligence index9.1→8.9-2.2%
ScoreQwen3 32B (Non-reasoning)AlibabaIntelligence index8.6→8.5-1.2%
ScoreQwen3 32B (Reasoning)AlibabaIntelligence index11.5→11.4-0.9%
ScoreQwen3 4B (Non-reasoning)AlibabaIntelligence index6.8→6.5-4.4%
ScoreQwen3 4B (Reasoning)AlibabaIntelligence index8.4→8.2-2.4%
ScoreQwen3 4B 2507 (Reasoning)AlibabaIntelligence index12→11.9-0.8%
ScoreQwen3 4B 2507 InstructAlibabaIntelligence index7.1→6.9-2.8%
ScoreQwen3 8B (Non-reasoning)AlibabaIntelligence index5.1→4.8-5.9%
ScoreQwen3 8B (Reasoning)AlibabaAgentic index1.5→1.6+6.7%
ScoreQwen3 Coder 480B A35B InstructAlibabaIntelligence index18→18.2+1.1%
ScoreQwen3 Coder NextAlibabaIntelligence index21.1→21.3+0.9%
ScoreQwen3 Coder NextAlibabaAgentic index8.8→8.9+1.1%
ScoreQwen3 MaxAlibabaIntelligence index24→24.5+2.1%
ScoreQwen3 Max (Preview)AlibabaIntelligence index19.2→19.4+1.0%
ScoreQwen3 Max ThinkingAlibabaIntelligence index31.7→32.5+2.5%
ScoreQwen3 Max Thinking (Preview)AlibabaIntelligence index25→25.5+2.0%
ScoreQwen3 Next 80B A3B (Reasoning)AlibabaIntelligence index16.7→16.9+1.2%
ScoreQwen3 Next 80B A3B InstructAlibabaIntelligence index13.7→13.8+0.7%
ScoreQwen3 Omni 30B A3B (Reasoning)AlibabaIntelligence index9.6→9.5-1.0%
ScoreQwen3 Omni 30B A3B InstructAlibabaIntelligence index5.1→4.8-5.9%
ScoreQwen3 VL 235B A22B (Reasoning)AlibabaIntelligence index20.6→20.9+1.5%
ScoreQwen3 VL 235B A22B InstructAlibabaIntelligence index14.3→14.4+0.7%
ScoreQwen3 VL 30B A3B (Reasoning)AlibabaIntelligence index13.3→13.4+0.8%
ScoreQwen3 VL 30B A3B InstructAlibabaIntelligence index10→9.9-1.0%
ScoreQwen3 VL 32B (Reasoning)AlibabaIntelligence index17.9→18.1+1.1%
ScoreQwen3 VL 32B InstructAlibabaIntelligence index11.1→11-0.9%
ScoreQwen3 VL 4B (Reasoning)AlibabaIntelligence index7.9→7.7-2.5%
ScoreQwen3 VL 4B InstructAlibabaIntelligence index4.1→3.7-9.8%
ScoreQwen3 VL 8B (Reasoning)AlibabaIntelligence index10.6→10.5-0.9%
ScoreQwen3 VL 8B InstructAlibabaIntelligence index8.4→8.2-2.4%
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaIntelligence index3.3→2.9-12%
ScoreQwen3.5 0.8B (Non-reasoning)AlibabaAgentic index0.4→Metric withdrawn
ScoreQwen3.5 0.8B (Reasoning)AlibabaIntelligence index5.5→5.2-5.5%
ScoreQwen3.5 0.8B (Reasoning)AlibabaAgentic index5.7→Metric withdrawn
ScoreQwen3.5 122B A10B (Non-reasoning)AlibabaIntelligence index27.6→28.2+2.2%
ScoreQwen3.5 122B A10B (Non-reasoning)AlibabaAgentic index15.8→16.2+2.5%
ScoreQwen3.5 122B A10B (Reasoning)AlibabaIntelligence index32.3→32.8+1.5%
ScoreQwen3.5 122B A10B (Reasoning)AlibabaAgentic index20.7→21.3+2.9%
ScoreQwen3.5 27B (Non-reasoning)AlibabaIntelligence index29.3→30+2.4%
ScoreQwen3.5 27B (Reasoning)AlibabaIntelligence index33.8→34.6+2.4%
ScoreQwen3.5 2B (Non-reasoning)AlibabaIntelligence index5.6→5.3-5.4%
ScoreQwen3.5 2B (Non-reasoning)AlibabaAgentic index0.8→Metric withdrawn
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index7.6→7.4-2.6%
ScoreQwen3.5 2B (Reasoning)AlibabaAgentic index0.8→Metric withdrawn
ScoreQwen3.5 35B A3B (Non-reasoning)AlibabaIntelligence index24→24.3+1.3%
ScoreQwen3.5 35B A3B (Reasoning)AlibabaIntelligence index29.3→29.9+2.0%
ScoreQwen3.5 397B A17B (Non-reasoning)AlibabaIntelligence index32→32.7+2.2%
ScoreQwen3.5 397B A17B (Reasoning)AlibabaIntelligence index33.7→34.3+1.8%
ScoreQwen3.5 4B (Non-reasoning)AlibabaIntelligence index16→16.1+0.6%
ScoreQwen3.5 4B (Reasoning)AlibabaIntelligence index20.1→20.4+1.5%
ScoreQwen3.5 9B (Non-reasoning)AlibabaIntelligence index20.3→20.6+1.5%
ScoreQwen3.5 9B (Reasoning)AlibabaIntelligence index21.4→21.8+1.9%
ScoreQwen3.5 9B (Reasoning)AlibabaAgentic index7.4→7-5.4%
ScoreQwen3.5 Omni FlashAlibabaIntelligence index19→19.2+1.1%
ScoreQwen3.5 Omni PlusAlibabaIntelligence index30.6→31.3+2.3%
ScoreQwen3.6 27B (Non-reasoning)AlibabaIntelligence index30.5→31.3+2.6%
ScoreQwen3.6 27B (Non-reasoning)AlibabaAgentic index23.3→23.6+1.3%
ScoreQwen3.6 27B (Reasoning)AlibabaIntelligence index37.1→37.7+1.6%
ScoreQwen3.6 27B (Reasoning)AlibabaAgentic index27→27.5+1.9%
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaIntelligence index24.2→24.6+1.7%
ScoreQwen3.6 35B A3B (Non-reasoning)AlibabaAgentic index19.1→Metric withdrawn
ScoreQwen3.6 35B A3B (Reasoning)AlibabaIntelligence index31.6→32.1+1.6%
ScoreQwen3.6 35B A3B (Reasoning)AlibabaAgentic index21.4→21.6+0.9%
ScoreQwen3.6 Max PreviewAlibabaIntelligence index40→41.1+2.8%
ScoreQwen3.6 PlusAlibabaIntelligence index39.6→40.5+2.3%
ScoreQwen3.6 PlusAlibabaAgentic index27.6→29+5.1%
ScoreQwen3.7 MaxAlibabaIntelligence index46→46.7+1.5%
ScoreQwen3.7 MaxAlibabaAgentic index30.6→30.9+1.0%
ScoreQwen3.7 PlusAlibabaIntelligence index39→39.4+1.0%
ScoreQwen3.7 PlusAlibabaAgentic index20.8→20.7-0.5%
ScoreQwQ 32B-PreviewAlibabaIntelligence index9.2→9.1-1.1%
ScoreR1 1776PerplexityIntelligence index6.3→6-4.8%
ScoreReka Flash (Sep '24)Reka AIIntelligence index6.3→6-4.8%
ScoreReka Flash 3Reka AIIntelligence index4.1→3.7-9.8%
ScoreRing-1TInclusionAIIntelligence index16.2→16.3+0.6%
ScoreRing-2.6-1TInclusionAIIntelligence index30.6→31.3+2.3%
ScoreRing-2.6-1TInclusionAIAgentic index18.9→Metric withdrawn
ScoreRing-flash-2.0InclusionAIIntelligence index8.2→8-2.4%
ScoreSarvam 30B (high)SarvamIntelligence index6.6→6.4-3.0%
ScoreSarvam M (Reasoning, based on Mistral Small 3.1)SarvamIntelligence index3→2.6-13%
ScoreSeed-OSS-36B-InstructByteDance SeedIntelligence index18.3→18.5+1.1%
ScoreSolar MiniUpstageIntelligence index6.2→6-3.2%
ScoreSolar Open 100B (Reasoning)UpstageIntelligence index15.1→15.2+0.7%
ScoreSolar Pro 2 (Non-reasoning)UpstageIntelligence index7.8→7.6-2.6%
ScoreSolar Pro 2 (Preview) (Non-reasoning)UpstageIntelligence index10→9.9-1.0%
ScoreSolar Pro 2 (Reasoning)UpstageIntelligence index9→8.8-2.2%
ScoreSolar Pro 3UpstageIntelligence index14.1→14.5+2.8%
ScoreSolar Pro 3UpstageAgentic index2.7→2.9+7.4%
ScoreSonarPerplexityIntelligence index9.5→9.4-1.1%
ScoreSonar ProPerplexityIntelligence index9.3→9.1-2.2%
ScoreSonar ReasoningPerplexityIntelligence index11.7→11.6-0.9%
ScoreSonar Reasoning ProPerplexityIntelligence index17.8→18+1.1%
ScoreStep 3.5 FlashStepFunIntelligence index25.5→26+2.0%
ScoreStep 3.5 Flash 2603StepFunIntelligence index26→26.5+1.9%
ScoreStep 3.7 FlashStepFunIntelligence index30.3→30.9+2.0%
ScoreStep 3.7 FlashStepFunAgentic index21.5→21.7+0.9%
ScoreStep3 VL 10BStepFunIntelligence index9.5→9.3-2.1%
ScoreTri-21B-ThinkTrillion LabsIntelligence index12.4→12.3-0.8%
ScoreTrinity Large ThinkingArcee AIIntelligence index18.2→18.4+1.1%
ScoreTrinity Large ThinkingArcee AIAgentic index3.7→Metric withdrawn

08-05

Aug 5, 202612 changes

RemovedQwen3.8 MaxAlibaba
PriceCeleris-1CelerisInput price$2.00→$0.200-90%
PriceCeleris-1CelerisOutput price$6.00→$0.700-88%
PriceGemma 4 31B (Non-reasoning)GoogleInput price$0.140→$0.150+7.1%
PriceGLM-5.2 (max)Z AIInput price$1.40→$1.35-3.6%
PriceGLM-5.2 (max)Z AIOutput price$4.40→$4.29-2.5%
PriceGrok 4.20 0309 v2 (Reasoning)SpaceXAIInput price$2.00→$1.25-38%
PriceGrok 4.20 0309 v2 (Reasoning)SpaceXAIOutput price$6.00→$2.50-58%
PriceHy3-preview (Non-reasoning)TencentOutput price$0.230→$0.210-8.7%
PriceHy3-preview (Reasoning)TencentOutput price$0.230→$0.210-8.7%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAInput price$0.680→$0.600-12%
PriceNemotron 3 Ultra 550B A55B (Reasoning)NVIDIAOutput price$2.67→$2.75+3.0%

08-04

Aug 4, 20267 changes

AddedG9v3-39A5BAI9StarsIntelligence index 30.9
AddedQwen3.8 MaxAlibabaIntelligence index 53.4
PriceGrok 4.20 0309 v2 (Non-reasoning)SpaceXAIInput price$2.00→$1.25-38%
PriceGrok 4.20 0309 v2 (Non-reasoning)SpaceXAIOutput price$6.00→$2.50-58%
PriceInkling (xhigh)Thinking MachinesInput price$1.87→$1.00-47%
PriceInkling (xhigh)Thinking MachinesOutput price$4.68→$4.05-13%
PriceLlama 4 MaverickMetaOutput price$0.850→$0.910+7.1%

08-01

Aug 1, 20261 changes

AddedCeleris-1CelerisIntelligence index 11.8

07-31

Jul 31, 202649 changes

AddedDeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekIntelligence index 49.9
AddedInkling SmallThinking MachinesIntelligence index 40.2
AddedKimi K3 (low)KimiIntelligence index 46.6
PriceGPT-5 (ChatGPT)OpenAIInput price$1.25→Metric withdrawn
PriceGPT-5 (ChatGPT)OpenAIOutput price$10.00→Metric withdrawn
PriceGPT-5.6 Luna (high)OpenAIInput price$1.00→$0.200-80%
PriceGPT-5.6 Luna (high)OpenAIOutput price$6.00→$1.20-80%
PriceGPT-5.6 Luna (low)OpenAIInput price$1.00→$0.200-80%
PriceGPT-5.6 Luna (low)OpenAIOutput price$6.00→$1.20-80%
PriceGPT-5.6 Luna (max)OpenAIInput price$1.00→$0.200-80%
PriceGPT-5.6 Luna (max)OpenAIOutput price$6.00→$1.20-80%
PriceGPT-5.6 Luna (medium)OpenAIInput price$1.00→$0.200-80%
PriceGPT-5.6 Luna (medium)OpenAIOutput price$6.00→$1.20-80%
PriceGPT-5.6 Luna (Non-reasoning)OpenAIInput price$1.00→$0.200-80%
PriceGPT-5.6 Luna (Non-reasoning)OpenAIOutput price$6.00→$1.20-80%
PriceGPT-5.6 Luna (xhigh)OpenAIInput price$1.00→$0.200-80%
PriceGPT-5.6 Luna (xhigh)OpenAIOutput price$6.00→$1.20-80%
PriceGPT-5.6 Terra (high)OpenAIInput price$2.50→$2.00-20%
PriceGPT-5.6 Terra (high)OpenAIOutput price$15.00→$12.00-20%
PriceGPT-5.6 Terra (low)OpenAIInput price$2.50→$2.00-20%
PriceGPT-5.6 Terra (low)OpenAIOutput price$15.00→$12.00-20%
PriceGPT-5.6 Terra (max)OpenAIInput price$2.50→$2.00-20%
PriceGPT-5.6 Terra (max)OpenAIOutput price$15.00→$12.00-20%
PriceGPT-5.6 Terra (medium)OpenAIInput price$2.50→$2.00-20%
PriceGPT-5.6 Terra (medium)OpenAIOutput price$15.00→$12.00-20%
PriceGPT-5.6 Terra (Non-reasoning)OpenAIInput price$2.50→$2.00-20%
PriceGPT-5.6 Terra (Non-reasoning)OpenAIOutput price$15.00→$12.00-20%
PriceGPT-5.6 Terra (xhigh)OpenAIInput price$2.50→$2.00-20%
PriceGPT-5.6 Terra (xhigh)OpenAIOutput price$15.00→$12.00-20%
Pricegpt-oss-120b (low)OpenAIOutput price$0.600→$0.590-1.7%
Pricegpt-oss-20b (high)OpenAIInput price$0.060→$0.070+17%
Pricegpt-oss-20b (low)OpenAIOutput price$0.200→$0.230+15%
PriceHy3TencentOutput price$0.580→$0.560-3.4%
PriceLlama 3.2 Instruct 11B (Vision)MetaInput price$0.360→$0.340-5.6%
PriceLlama 3.2 Instruct 11B (Vision)MetaOutput price$0.360→$0.340-5.6%
PriceLlama 3.2 Instruct 90B (Vision)MetaInput price$2.04→Metric withdrawn
PriceLlama 3.2 Instruct 90B (Vision)MetaOutput price$2.04→Metric withdrawn
PriceLlama 3.3 Instruct 70BMetaInput price$0.590→$0.600+1.7%
PriceLongCat 2.0LongCatInput priceFirst measured→$0.750
PriceLongCat 2.0LongCatOutput priceFirst measured→$2.95
PriceNemotron 3 Super 120B A12B (Reasoning)NVIDIAInput price$0.250→$0.200-20%
PriceNemotron 3 Super 120B A12B (Reasoning)NVIDIAOutput price$0.780→$0.800+2.6%
PriceSolar Pro 3UpstageInput priceFirst measured→$0.150
PriceSolar Pro 3UpstageOutput priceFirst measured→$0.600
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index9.4→9.7+3.2%
ScoreGemma 4 E4B (Reasoning)GoogleIntelligence index11.9→12.2+2.5%
ScoreMinistral 3 3BMistralIntelligence index6.3→6.8+7.9%
ScoreQwen3.5 0.8B (Reasoning)AlibabaIntelligence index5.3→5.5+3.8%
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index7.1→7.6+7.0%

07-28

Jul 28, 202614 changes

PriceGLM-5.1 (Reasoning)Z AIInput price$1.39→$1.38-0.7%
PriceHy3-preview (Non-reasoning)TencentInput price$0.120→$0.060-50%
PriceHy3-preview (Non-reasoning)TencentOutput price$0.430→$0.230-47%
PriceHy3-preview (Reasoning)TencentInput price$0.120→$0.060-50%
PriceHy3-preview (Reasoning)TencentOutput price$0.430→$0.230-47%
PriceLlama 3.3 Instruct 70BMetaOutput price$0.710→$0.720+1.4%
PriceMistral MediumMistralInput price$2.75→$1.50-45%
PriceMistral MediumMistralOutput price$8.10→$7.50-7.4%
PriceMistral Small (Feb '24)MistralInput price$1.00→$0.150-85%
PriceMistral Small (Feb '24)MistralOutput price$3.00→$0.600-80%
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index9.5→9.4-1.1%
ScoreMinistral 3 3BMistralIntelligence index6.5→6.3-3.1%
ScoreQwen3.5 0.8B (Reasoning)AlibabaIntelligence index5.5→5.3-3.6%
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index6.9→7.1+2.9%

07-25

Jul 25, 202628 changes

AddedAgnes 2.5 Pro AlphaSapiens AIIntelligence index 38.8
AddedClaude Opus 5 (Adaptive Reasoning, High Effort)AnthropicIntelligence index 58.9
AddedClaude Opus 5 (Adaptive Reasoning, Low Effort)AnthropicIntelligence index 50.6
AddedClaude Opus 5 (Adaptive Reasoning, Max Effort)AnthropicIntelligence index 60.7
AddedClaude Opus 5 (Adaptive Reasoning, Medium Effort)AnthropicIntelligence index 56.3
AddedClaude Opus 5 (Adaptive Reasoning, Xhigh Effort)AnthropicIntelligence index 60.1
PriceGemma 3n E4B InstructGoogleInput price$0.020→$0.060+200%
PriceGemma 3n E4B InstructGoogleOutput price$0.040→$0.120+200%
PriceGLM-5.1 (Non-reasoning)Z AIInput price$1.40→$1.38-1.4%
PriceGLM-5.1 (Reasoning)Z AIInput price$1.40→$1.39-0.7%
PriceGLM-5.2 (Non-reasoning)Z AIInput price$1.40→$1.35-3.6%
PriceGLM-5.2 (Non-reasoning)Z AIOutput price$4.40→$4.25-3.4%
PriceGrok 4SpaceXAIInput price$5.50→$3.00-45%
PriceGrok 4SpaceXAIOutput price$27.50→$15.00-45%
PriceLlama 4 MaverickMetaInput price$0.310→$0.270-13%
PriceMiMo-V2.5-Pro (Non-reasoning)XiaomiInput price$0.660→$0.430-35%
PriceMiMo-V2.5-Pro (Non-reasoning)XiaomiOutput price$1.72→$0.870-49%
PriceQwen3.5 9B (Reasoning)AlibabaInput price$0.100→$0.140+40%
PriceQwen3.5 9B (Reasoning)AlibabaOutput price$0.150→$0.200+33%
ScoreEXAONE 4.5 33BLG AI ResearchIntelligence index23→20.2-12%
ScoreEXAONE 4.5 33BLG AI ResearchCoding indexFirst measured→23.6
ScoreEXAONE 4.5 33BLG AI ResearchAgentic indexFirst measured→9.7
ScoreGemma 4 12B (Reasoning)GoogleIntelligence index22→21.8-0.9%
ScoreGemma 4 12B (Reasoning)GoogleCoding indexFirst measured→31
ScoreGemma 4 E2B (Reasoning)GoogleIntelligence index9.3→9.5+2.2%
ScoreGemma 4 E4B (Reasoning)GoogleIntelligence index12.2→11.9-2.5%
ScoreMinistral 3 3BMistralIntelligence index6.8→6.5-4.4%
ScoreQwen3.5 2B (Reasoning)AlibabaIntelligence index7.6→6.9-9.2%

07-24

Jul 24, 20267 changes

AddedG9v3-3BAI9StarsIntelligence index 16.1
ScoreK-EXAONE (Reasoning)LG AI ResearchIntelligence index24.7→22.1-11%
ScoreK-EXAONE (Reasoning)LG AI ResearchCoding indexFirst measured→32.1
ScoreK-EXAONE (Reasoning)LG AI ResearchAgentic indexFirst measured→8
ScoreNemotron Cascade 2 30B A3BNVIDIAIntelligence index21.3→17.6-17%
ScoreNemotron Cascade 2 30B A3BNVIDIACoding indexFirst measured→25.3
ScoreNemotron Cascade 2 30B A3BNVIDIAAgentic indexFirst measured→3.7