2,339
变化
62 天记录每天与前一日快照对比,记录新增、下架、分数与价格的变动。
2026年7月23日 — 2026年9月26日
09-26
2026年9月26日1 条变化- 分数Apodex 1.1Apodex智能指数30.4→26.4-13%
09-25
2026年9月25日9 条变化- 新增DeepSeek V4.1 Flash (Non-Reasoning)DeepSeek智能指数 24.7
- 新增GLM-5.3 (low)Z AI智能指数 34.3
- 下架Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B)Sapiens AI
- 下架Agnes 2.5 Pro BetaSapiens AI
- 下架Agnes 3.0 FlashSapiens AI
- 价格Gemma 4 26B A4B (Reasoning)Google输入价$0.120→$0.100-17%
- 价格Muse Glimmer (high)Meta输入价$0.350→$0.320-8.6%
- 价格Muse Glimmer (high)Meta输出价$1.50→$1.35-10.0%
- 分数Apodex 1.1Apodex智能体指数首次评测→28.4
09-24
2026年9月24日36 条变化- 新增Mercury 2.5Inception智能指数 12.3
- 价格Devstral 2Mistral输入价$0→指标撤下
- 价格Devstral 2Mistral输出价$0→指标撤下
- 价格Devstral Small 2Mistral输入价$0→指标撤下
- 价格Devstral Small 2Mistral输出价$0→指标撤下
- 价格LFM2.5-8B-A1BLiquid AI输入价$0→指标撤下
- 价格LFM2.5-8B-A1BLiquid AI输出价$0→指标撤下
- 价格LFM2.5-VL-1.6BLiquid AI输入价$0→指标撤下
- 价格LFM2.5-VL-1.6BLiquid AI输出价$0→指标撤下
- 价格Ling-3.0-flash-FinInclusionAI输入价$0→$0.070—
- 价格Ling-3.0-flash-FinInclusionAI输出价$0→$0.220—
- 价格Llama 3.1 Nemotron Instruct 70BNVIDIA输入价$1.20→指标撤下
- 价格Llama 3.1 Nemotron Instruct 70BNVIDIA输出价$1.20→指标撤下
- 价格Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA输入价$0.400→指标撤下
- 价格Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA输出价$0.400→指标撤下
- 价格Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA输入价$0.400→指标撤下
- 价格Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA输出价$0.400→指标撤下
- 价格Mistral 7B InstructMistral输入价$0.250→$0.150-40%
- 价格Mistral 7B InstructMistral输出价$0.250→$0.200-20%
- 价格Mistral MediumMistral输入价$1.50→指标撤下
- 价格Mistral MediumMistral输出价$7.50→指标撤下
- 价格Mistral Medium 3.1Mistral输入价$0.400→指标撤下
- 价格Mistral Medium 3.1Mistral输出价$2.00→指标撤下
- 价格Mistral Small (Feb '24)Mistral输入价$0.150→指标撤下
- 价格Mistral Small (Feb '24)Mistral输出价$0.600→指标撤下
- 价格Mistral Small (Sep '24)Mistral输入价$0.200→指标撤下
- 价格Mistral Small (Sep '24)Mistral输出价$0.600→指标撤下
- 价格Mistral Small 3Mistral输入价$0.100→$0.050-50%
- 价格Mistral Small 3Mistral输出价$0.300→$0.080-73%
- 价格Mistral Small 3.1Mistral输入价$0.100→$0.110+10.0%
- 价格Mistral Small 3.1Mistral输出价$0.300→$0.170-43%
- 价格Mistral Small 3.2Mistral输入价$0.100→$0.070-30%
- 价格Mistral Small 3.2Mistral输出价$0.300→$0.200-33%
- 价格NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA输入价$0.200→指标撤下
- 价格NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA输出价$0.600→指标撤下
- 价格Trinity Large ThinkingArcee AI输出价$0.900→$0.800-11%
09-23
2026年9月23日27 条变化- 新增Claude Opus 5.5 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic智能指数 53.6
- 新增Claude Opus 5.5 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic智能指数 42.3
- 新增Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic智能指数 57.6
- 新增Claude Opus 5.5 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic智能指数 51.2
- 新增Claude Opus 5.5 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic智能指数 56
- 新增GPT-6 Luna (high)OpenAI智能指数 32.1
- 新增GPT-6 Luna (low)OpenAI智能指数 20.9
- 新增GPT-6 Luna (max)OpenAI智能指数 37.3
- 新增GPT-6 Luna (medium)OpenAI智能指数 29.5
- 新增GPT-6 Luna (Non-reasoning)OpenAI智能指数 18.3
- 新增GPT-6 Luna (xhigh)OpenAI智能指数 33.9
- 新增GPT-6 Sol (high)OpenAI智能指数 42.8
- 新增GPT-6 Sol (low)OpenAI智能指数 33.9
- 新增GPT-6 Sol (max)OpenAI智能指数 47.5
- 新增GPT-6 Sol (medium)OpenAI智能指数 39.8
- 新增GPT-6 Sol (Non-reasoning)OpenAI智能指数 28.1
- 新增GPT-6 Sol (xhigh)OpenAI智能指数 44.1
- 价格GLM-5.1 (Reasoning)Z AI输入价$1.20→$1.28+6.7%
- 价格GLM-5.1 (Reasoning)Z AI输出价$4.40→$4.07-7.5%
- 价格gpt-oss-20b (high)OpenAI输入价$0.060→$0.070+17%
- 价格gpt-oss-20b (high)OpenAI输出价$0.190→$0.180-5.3%
- 价格gpt-oss-20b (low)OpenAI输出价$0.200→$0.210+5.0%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输出价$2.40→$2.50+4.2%
- 价格Nemotron 3.5 LightningNVIDIA输入价$0.060→$0.070+17%
- 价格Nemotron 3.5 LightningNVIDIA输出价$0.200→$0.220+10.0%
- 价格Nova PremierAmazon输入价$2.50→指标撤下
- 价格Nova PremierAmazon输出价$12.50→指标撤下
09-22
2026年9月22日10 条变化- 新增Grok 4.7 (high)SpaceXAI智能指数 46.3
- 新增Grok 4.7 (xhigh)SpaceXAI智能指数 46.4
- 新增MiMo-V2.6-ProXiaomi智能指数 46.3
- 价格Ling-3.0-flash-VLInclusionAI输入价$0→$0.070—
- 价格Ling-3.0-flash-VLInclusionAI输出价$0→$0.220—
- 价格Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA输入价$0.090→$0.200+122%
- 价格Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA输出价$0.360→$1.09+203%
- 价格Nemotron 3 Super 120B A12B (Reasoning)NVIDIA输入价$0.190→$0.300+58%
- 价格Nemotron 3 Super 120B A12B (Reasoning)NVIDIA输出价$0.650→$0.900+38%
- 分数Qwen3.8 27B (Non-reasoning)Alibaba智能指数22.4→20.2-9.8%
09-20
2026年9月20日277 条变化- 新增Step 5 PreviewStepFun智能指数 43.7
- 分数Agnes 2.5 Pro AlphaSapiens AI智能指数27.8→26.8-3.6%
- 分数Agnes 2.5 Pro AlphaSapiens AI智能体指数29.4→指标撤下
- 分数Claude 4.5 Haiku (Reasoning)Anthropic智能指数17.6→16.9-4.0%
- 分数Claude 4.5 Haiku (Reasoning)Anthropic智能体指数10.3→8-22%
- 分数Claude 4.5 Sonnet (Reasoning)Anthropic智能指数21.2→20.7-2.4%
- 分数Claude 4.5 Sonnet (Reasoning)Anthropic智能体指数17.5→15.8-9.7%
- 分数Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic智能指数49.7→49.6-0.2%
- 分数Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic智能体指数51→50.7-0.6%
- 分数Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic智能体指数53.3→53.1-0.4%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic智能指数47→46.8-0.4%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic智能体指数47.8→47.1-1.5%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic智能体指数58→57.9-0.2%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic智能指数49.1→48.9-0.4%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic智能体指数50.6→50.2-0.8%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic智能体指数57.1→57.2+0.2%
- 分数Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic智能体指数39.5→38.6-2.3%
- 分数Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能指数42→41.8-0.5%
- 分数Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能体指数42.6→41.9-1.6%
- 分数Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能指数48.2→48.1-0.2%
- 分数Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能体指数52.7→52.3-0.8%
- 分数Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能指数39.8→39.4-1.0%
- 分数Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能体指数37.5→36-4.0%
- 分数Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数50.7→50.8+0.2%
- 分数Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能体指数56.2→56.5+0.5%
- 分数Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能指数45.1→44.8-0.7%
- 分数Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能体指数47→46.2-1.7%
- 分数Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能体指数55.5→55.6+0.2%
- 分数Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能指数30.5→30.1-1.3%
- 分数Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能体指数33.1→31.8-3.9%
- 分数Claude Sonnet 5 (Adaptive Reasoning, High Effort)Anthropic智能指数32→31.7-0.9%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Low Effort)Anthropic智能指数24.7→24.3-1.6%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数38.4→38.2-0.5%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic智能体指数44.3→43.6-1.6%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)Anthropic智能指数28.4→28.1-1.1%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能指数34.7→34.4-0.9%
- 分数Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic智能指数28.9→23.2-20%
- 分数Command A+Cohere智能指数13.9→13.1-5.8%
- 分数Command A+Cohere智能体指数3.6→1-72%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能指数15.4→14.8-3.9%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能体指数8.9→6.8-24%
- 分数DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek智能指数24.8→26+4.8%
- 分数DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek智能体指数27.9→指标撤下
- 分数DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek智能指数24.6→24.2-1.6%
- 分数DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek智能体指数23.7→22.2-6.3%
- 分数DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能指数34.5→34.3-0.6%
- 分数DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能体指数41.7→41-1.7%
- 分数DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek智能指数35→34.8-0.6%
- 分数DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek智能体指数48.1→47.5-1.2%
- 分数DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek智能指数30.9→30.4-1.6%
- 分数DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek智能体指数27.7→26.3-5.1%
- 分数DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek智能指数36.3→36-0.8%
- 分数DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek智能体指数42.3→41.3-2.4%
- 分数Devstral 2Mistral智能指数9.4→8.6-8.5%
- 分数Devstral 2Mistral智能体指数4.9→2.3-53%
- 分数Devstral Small 2Mistral智能指数8.4→7.5-11%
- 分数Devstral Small 2Mistral智能体指数4.8→2-58%
- 分数Gemini 2.5 ProGoogle智能指数16.7→16.1-3.6%
- 分数Gemini 2.5 ProGoogle智能体指数3.5→1.6-54%
- 分数Gemini 3.1 Flash-LiteGoogle智能指数16→15.6-2.5%
- 分数Gemini 3.1 Flash-LiteGoogle智能体指数3.2→1.6-50%
- 分数Gemini 3.1 Pro PreviewGoogle智能指数30.4→29.7-2.3%
- 分数Gemini 3.1 Pro PreviewGoogle智能体指数10.3→8.2-20%
- 分数Gemini 3.5 Flash (high)Google智能指数33→32.6-1.2%
- 分数Gemini 3.5 Flash (high)Google智能体指数27.3→26-4.8%
- 分数Gemini 3.5 Flash-LiteGoogle智能指数22.7→22.2-2.2%
- 分数Gemini 3.5 Flash-LiteGoogle智能体指数15.9→14.3-10%
- 分数Gemini 3.6 Flash (high)Google智能指数34.3→34-0.9%
- 分数Gemini 3.6 Flash (high)Google智能体指数30.2→29-4.0%
- 分数Gemini 3.7 Flash (high)Google智能指数39.4→39.1-0.8%
- 分数Gemini 3.7 Flash (high)Google智能体指数36.4→35.2-3.3%
- 分数Gemini 3.8 Flash (high)Google智能指数41.2→40.9-0.7%
- 分数Gemini 3.8 Flash (high)Google智能体指数41.1→40.2-2.2%
- 分数Gemini 3.8 Flash (low)Google智能指数33.8→33.5-0.9%
- 分数Gemini 3.8 Flash (low)Google智能体指数33.7→32.6-3.3%
- 分数Gemini 3.8 Flash (medium)Google智能指数40→39.8-0.5%
- 分数Gemini 3.8 Flash (medium)Google智能体指数40.5→39.7-2.0%
- 分数Gemma 4 31B (Reasoning)Google智能指数15.4→19+23%
- 分数Gemma 4 31B (Reasoning)Google智能体指数6.7→4.2-37%
- 分数GLM 5.3 FlashZ AI智能指数41.9→41.8-0.2%
- 分数GLM 5.3 FlashZ AI智能体指数51.2→50.9-0.6%
- 分数GLM-5.1 (Reasoning)Z AI智能指数26.4→26.1-1.1%
- 分数GLM-5.1 (Reasoning)Z AI智能体指数25.2→23.9-5.2%
- 分数GLM-5.2 (max)Z AI智能指数34→33.7-0.9%
- 分数GLM-5.2 (max)Z AI智能体指数39.4→38.4-2.5%
- 分数GLM-5.3 (max)Z AI智能指数44.9→44.8-0.2%
- 分数GLM-5.3 (max)Z AI智能体指数53.4→53.1-0.6%
- 分数GPT-5 mini (high)OpenAI智能指数17.4→16.8-3.4%
- 分数GPT-5 mini (high)OpenAI智能体指数8.9→6.8-24%
- 分数GPT-5.4 mini (xhigh)OpenAI智能指数24.6→24.1-2.0%
- 分数GPT-5.4 mini (xhigh)OpenAI智能体指数19.7→17.9-9.1%
- 分数GPT-5.4 nano (xhigh)OpenAI智能指数21.2→20.7-2.4%
- 分数GPT-5.4 nano (xhigh)OpenAI智能体指数17.7→16-9.6%
- 分数GPT-5.5 (high)OpenAI智能指数37.3→37-0.8%
- 分数GPT-5.5 (high)OpenAI智能体指数35.4→34.3-3.1%
- 分数GPT-5.5 (medium)OpenAI智能指数34.2→33.8-1.2%
- 分数GPT-5.5 (medium)OpenAI智能体指数30.6→29.5-3.6%
- 分数GPT-5.5 (xhigh)OpenAI智能指数38.6→38.4-0.5%
- 分数GPT-5.5 (xhigh)OpenAI智能体指数37.3→36.4-2.4%
- 分数GPT-5.5 Instant (June 2026)OpenAI智能指数26.8→26-3.0%
- 分数GPT-5.5 Instant (June 2026)OpenAI智能体指数19.2→16.6-14%
- 分数GPT-5.6 Luna (high)OpenAI智能指数32.4→32.1-0.9%
- 分数GPT-5.6 Luna (high)OpenAI智能体指数35.6→34.6-2.8%
- 分数GPT-5.6 Luna (low)OpenAI智能指数21.5→21-2.3%
- 分数GPT-5.6 Luna (low)OpenAI智能体指数17.9→16.1-10%
- 分数GPT-5.6 Luna (max)OpenAI智能指数37.5→37.3-0.5%
- 分数GPT-5.6 Luna (max)OpenAI智能体指数42.7→42.1-1.4%
- 分数GPT-5.6 Luna (medium)OpenAI智能指数25.5→25-2.0%
- 分数GPT-5.6 Luna (medium)OpenAI智能体指数25.4→23.9-5.9%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能指数16.1→15.5-3.7%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能体指数15.2→13.4-12%
- 分数GPT-5.6 Luna (xhigh)OpenAI智能指数34.8→34.6-0.6%
- 分数GPT-5.6 Luna (xhigh)OpenAI智能体指数39.5→38.7-2.0%
- 分数GPT-5.6 Sol (high)OpenAI智能指数42.5→42.3-0.5%
- 分数GPT-5.6 Sol (high)OpenAI智能体指数44.7→44.2-1.1%
- 分数GPT-5.6 Sol (low)OpenAI智能指数33.8→33.5-0.9%
- 分数GPT-5.6 Sol (low)OpenAI智能体指数32.6→31.6-3.1%
- 分数GPT-5.6 Sol (max)OpenAI智能指数47.1→47-0.2%
- 分数GPT-5.6 Sol (max)OpenAI智能体指数50.5→50.2-0.6%
- 分数GPT-5.6 Sol (medium)OpenAI智能指数39.5→39.2-0.8%
- 分数GPT-5.6 Sol (medium)OpenAI智能体指数40.5→39.7-2.0%
- 分数GPT-5.6 Sol (Non-reasoning)OpenAI智能体指数29.4→28.1-4.4%
- 分数GPT-5.6 Sol (xhigh)OpenAI智能指数44.1→44-0.2%
- 分数GPT-5.6 Sol (xhigh)OpenAI智能体指数47.8→47.4-0.8%
- 分数GPT-5.6 Terra (high)OpenAI智能指数34.5→34.2-0.9%
- 分数GPT-5.6 Terra (high)OpenAI智能体指数37.6→36.7-2.4%
- 分数GPT-5.6 Terra (low)OpenAI智能指数27.9→27.5-1.4%
- 分数GPT-5.6 Terra (low)OpenAI智能体指数27.3→25.9-5.1%
- 分数GPT-5.6 Terra (max)OpenAI智能指数42.3→42.1-0.5%
- 分数GPT-5.6 Terra (max)OpenAI智能体指数43.7→43.2-1.1%
- 分数GPT-5.6 Terra (medium)OpenAI智能指数30.4→30.1-1.0%
- 分数GPT-5.6 Terra (medium)OpenAI智能体指数31.4→30.4-3.2%
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能指数21.2→20.8-1.9%
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能体指数24.9→23.5-5.6%
- 分数GPT-5.6 Terra (xhigh)OpenAI智能指数38.2→38-0.5%
- 分数GPT-5.6 Terra (xhigh)OpenAI智能体指数42.2→41.4-1.9%
- 分数GPT-6 Astra (high)OpenAI智能指数51→50.9-0.2%
- 分数GPT-6 Astra (high)OpenAI智能体指数48.7→48.2-1.0%
- 分数GPT-6 Astra (low)OpenAI智能指数46→45.8-0.4%
- 分数GPT-6 Astra (low)OpenAI智能体指数39.5→38.8-1.8%
- 分数GPT-6 Astra (max)OpenAI智能指数52.8→52.7-0.2%
- 分数GPT-6 Astra (max)OpenAI智能体指数51.5→51-1.0%
- 分数GPT-6 Astra (medium)OpenAI智能指数49.7→49.6-0.2%
- 分数GPT-6 Astra (medium)OpenAI智能体指数46.3→46-0.6%
- 分数GPT-6 Astra (xhigh)OpenAI智能指数52.5→52.4-0.2%
- 分数GPT-6 Astra (xhigh)OpenAI智能体指数50.6→50.2-0.8%
- 分数gpt-oss-120b (high)OpenAI智能指数12.3→11.6-5.7%
- 分数gpt-oss-120b (high)OpenAI智能体指数6.2→3.7-40%
- 分数gpt-oss-20b (high)OpenAI智能体指数1.4→1.2-14%
- 分数Granite 4.2 8BIBM智能指数11.8→11.1-5.9%
- 分数Granite 4.2 8BIBM智能体指数3.7→1.3-65%
- 分数Grok 4.3 (high)SpaceXAI智能指数25.4→24.9-2.0%
- 分数Grok 4.3 (high)SpaceXAI智能体指数17.2→15.5-9.9%
- 分数Grok 4.3 (Non-reasoning)SpaceXAI智能指数14.5→14-3.4%
- 分数Grok 4.3 (Non-reasoning)SpaceXAI智能体指数15.8→14-11%
- 分数Grok 4.5 (high)SpaceXAI智能指数39.1→38.8-0.8%
- 分数Grok 4.5 (high)SpaceXAI智能体指数42.1→41.2-2.1%
- 分数Grok 4.6 (high)SpaceXAI智能指数44.4→44.3-0.2%
- 分数Grok 4.6 (high)SpaceXAI智能体指数53.4→53-0.7%
- 分数Grok 4.6 (low)SpaceXAI智能指数35.4→35.1-0.8%
- 分数Grok 4.6 (low)SpaceXAI智能体指数42.2→41.2-2.4%
- 分数Grok 4.6 (medium)SpaceXAI智能指数43→42.8-0.5%
- 分数Grok 4.6 (medium)SpaceXAI智能体指数51.2→50.6-1.2%
- 分数Grok 4.6 (xhigh)SpaceXAI智能指数44.3→44.2-0.2%
- 分数Grok 4.6 (xhigh)SpaceXAI智能体指数52.7→52.5-0.4%
- 分数Hy3Tencent智能指数25.8→25.3-1.9%
- 分数Hy3Tencent智能体指数25.6→24.1-5.9%
- 分数HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing智能体指数2.7→0.9-67%
- 分数Inkling (xhigh)Thinking Machines智能指数25.5→25-2.0%
- 分数Inkling (xhigh)Thinking Machines智能体指数24.3→22.5-7.4%
- 分数Inkling SmallThinking Machines智能指数26.1→27.8+6.5%
- 分数Inkling SmallThinking Machines智能体指数25→指标撤下
- 分数K2 Horizon 3.7BInstitute of Foundation Models智能指数16.2→15.6-3.7%
- 分数K2 Horizon 3.7BInstitute of Foundation Models智能体指数18.4→16.5-10%
- 分数K2 Horizon 375B A23BInstitute of Foundation Models智能指数30.8→30.5-1.0%
- 分数K2 Horizon 375B A23BInstitute of Foundation Models智能体指数40.9→40-2.2%
- 分数K2 Horizon 7BInstitute of Foundation Models智能指数21→20.6-1.9%
- 分数K2 Horizon 7BInstitute of Foundation Models智能体指数27.6→26.1-5.4%
- 分数K2 Horizon MoVA 36B A4BInstitute of Foundation Models智能指数25.7→25.3-1.6%
- 分数K2 Horizon MoVA 36B A4BInstitute of Foundation Models智能体指数34.7→33.5-3.5%
- 分数Kimi K2.6Kimi智能指数27.5→27-1.8%
- 分数Kimi K2.6Kimi智能体指数22.1→20.5-7.2%
- 分数Kimi K2.7 CodeKimi智能指数26.3→25.8-1.9%
- 分数Kimi K2.7 CodeKimi智能体指数22.5→21-6.7%
- 分数Kimi K3 (low)Kimi智能指数30.5→34.5+13%
- 分数Kimi K3 (low)Kimi智能体指数30.7→指标撤下
- 分数Kimi K3 (max)Kimi智能指数43.8→43.6-0.5%
- 分数Kimi K3 (max)Kimi智能体指数50.6→50-1.2%
- 分数Ling 3.0 FlashInclusionAI智能指数20.6→24.9+21%
- 分数Ling 3.0 FlashInclusionAI智能体指数21→指标撤下
- 分数Ling 3.0 TinyInclusionAI智能指数11.9→15.3+29%
- 分数Ling 3.0 TinyInclusionAI智能体指数7.1→指标撤下
- 分数Ling-3.0-flash-FinInclusionAI智能指数23→22.6-1.7%
- 分数Ling-3.0-flash-FinInclusionAI智能体指数29.3→27.9-4.8%
- 分数Ling-3.0-flash-VLInclusionAI智能指数25→24.6-1.6%
- 分数Ling-3.0-flash-VLInclusionAI智能体指数30→28.7-4.3%
- 分数Llama 4 MaverickMeta智能指数9.3→10+7.5%
- 分数Llama 4 ScoutMeta智能指数6.5→8.1+25%
- 分数LongCat 2.0LongCat智能指数19.7→19.1-3.0%
- 分数LongCat 2.0LongCat智能体指数15.9→14-12%
- 分数Mercury 2Inception智能指数11.5→13.8+20%
- 分数Mercury 2Inception智能体指数4→1.6-60%
- 分数MiMo-V2.5Xiaomi智能指数22.3→25.2+13%
- 分数MiMo-V2.5Xiaomi智能体指数17.4→15.8-9.2%
- 分数MiMo-V2.5-ProXiaomi智能指数26.4→26-1.5%
- 分数MiMo-V2.5-ProXiaomi智能体指数22.7→21.3-6.2%
- 分数MiniCPM5-2BOpenBMB智能指数13.1→12.5-4.6%
- 分数MiniCPM5-2BOpenBMB智能体指数9→6.7-26%
- 分数MiniMax-M2.7MiniMax智能指数23.2→22.8-1.7%
- 分数MiniMax-M2.7MiniMax智能体指数16.8→15.3-8.9%
- 分数MiniMax-M3MiniMax智能指数29.6→29.2-1.4%
- 分数MiniMax-M3MiniMax智能体指数30.8→29.5-4.2%
- 分数Mistral Large 3Mistral智能指数9.7→9.3-4.1%
- 分数Mistral Large 3Mistral智能体指数2.4→1-58%
- 分数Mistral Medium 3.1Mistral智能指数9.5→9.2-3.2%
- 分数Mistral Medium 3.1Mistral智能体指数3.1→1.9-39%
- 分数Mistral Medium 3.5Mistral智能指数14.9→14.2-4.7%
- 分数Mistral Medium 3.5Mistral智能体指数9.4→7-26%
- 分数Mistral Small 3.1Mistral智能指数7.4→7.1-4.1%
- 分数Mistral Small 3.1Mistral智能体指数2.1→1.2-43%
- 分数Mistral Small 3.2Mistral智能指数7→8.2+17%
- 分数Mistral Small 4 (Reasoning)Mistral智能指数11.5→11.3-1.7%
- 分数Mistral Small 4 (Reasoning)Mistral智能体指数1.4→0.8-43%
- 分数Muse Glimmer (high)Meta智能指数18.1→17.5-3.3%
- 分数Muse Glimmer (high)Meta智能体指数10.5→8.5-19%
- 分数Muse Spark 1.1 (xhigh)Meta智能指数34.3→33.7-1.7%
- 分数Muse Spark 1.1 (xhigh)Meta智能体指数27.5→25.8-6.2%
- 分数Muse Spark 1.2 (xhigh)Meta智能指数39.8→39.6-0.5%
- 分数Muse Spark 1.2 (xhigh)Meta智能体指数44→43.2-1.8%
- 分数Muse Spark 1.3 (max)Meta智能指数48.2→48.1-0.2%
- 分数Muse Spark 1.3 (max)Meta智能体指数55.7→55.5-0.4%
- 分数Muse Spark 1.3 (xhigh)Meta智能指数45.2→45.1-0.2%
- 分数Muse Spark 1.3 (xhigh)Meta智能体指数51.8→51.5-0.6%
- 分数Nemotron 3 Super 120B A12B (Reasoning)NVIDIA智能指数13.6→12.8-5.9%
- 分数Nemotron 3 Super 120B A12B (Reasoning)NVIDIA智能体指数4.1→1.7-59%
- 分数Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA智能指数23.4→22.9-2.1%
- 分数Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA智能体指数21.7→20.1-7.4%
- 分数Nemotron 3.5 LightningNVIDIA智能指数13.6→12.9-5.1%
- 分数Nemotron 3.5 LightningNVIDIA智能体指数6.1→3.5-43%
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能体指数首次评测→23.7
- 分数Quasar 438B (max, based on GLM-5.2)Multiverse Computing智能指数27.1→26.7-1.5%
- 分数Quasar 438B (max, based on GLM-5.2)Multiverse Computing智能体指数32.7→31.4-4.0%
- 分数Qwen3 14B (Reasoning)Alibaba智能指数6.4→8.2+28%
- 分数Qwen3 32B (Reasoning)Alibaba智能指数7.2→8.6+19%
- 分数Qwen3 8B (Reasoning)Alibaba智能指数5.2→7.3+40%
- 分数Qwen3 Coder NextAlibaba智能指数10.1→9.2-8.9%
- 分数Qwen3 Coder NextAlibaba智能体指数3.6→0.9-75%
- 分数Qwen3.5 122B A10B (Reasoning)Alibaba智能指数16.2→15.6-3.7%
- 分数Qwen3.5 122B A10B (Reasoning)Alibaba智能体指数9.6→7.6-21%
- 分数Qwen3.5 397B A17B (Reasoning)Alibaba智能指数19.1→18.4-3.7%
- 分数Qwen3.5 397B A17B (Reasoning)Alibaba智能体指数10.6→8.3-22%
- 分数Qwen3.6 27B (Reasoning)Alibaba智能指数21.9→21.4-2.3%
- 分数Qwen3.6 27B (Reasoning)Alibaba智能体指数20.1→18.5-8.0%
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能指数18.8→18.2-3.2%
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能体指数15→13.1-13%
- 分数Qwen3.7 MaxAlibaba智能指数29.9→29.5-1.3%
- 分数Qwen3.7 MaxAlibaba智能体指数23.9→22.5-5.9%
- 分数Qwen3.7 PlusAlibaba智能指数25.8→25.2-2.3%
- 分数Qwen3.7 PlusAlibaba智能体指数19.7→17.5-11%
- 分数Qwen3.8 2.4T A95BAlibaba智能指数40→39.9-0.3%
- 分数Qwen3.8 2.4T A95BAlibaba智能体指数50.4→50.1-0.6%
- 分数Qwen3.8 27B (low)Alibaba智能指数26.5→26.2-1.1%
- 分数Qwen3.8 27B (low)Alibaba智能体指数39.9→38.9-2.5%
- 分数Qwen3.8 27B (medium)Alibaba智能指数27.8→27.6-0.7%
- 分数Qwen3.8 27B (medium)Alibaba智能体指数45.3→44.4-2.0%
- 分数Qwen3.8 27B (Non-reasoning)Alibaba智能体指数首次评测→22.4
- 分数Qwen3.8 27B (xhigh)Alibaba智能指数33.9→33.7-0.6%
- 分数Qwen3.8 27B (xhigh)Alibaba智能体指数46.5→45.8-1.5%
- 分数Qwen3.8 MaxAlibaba智能指数40.3→40.2-0.2%
- 分数Qwen3.8 MaxAlibaba智能体指数49.6→49.1-1.0%
- 分数Qwen3.8 Max (0902)Alibaba智能体指数56.1→56-0.2%
- 分数Qwen3.8-Flash-NextAlibaba智能指数39.9→39.8-0.3%
- 分数Qwen3.8-Flash-NextAlibaba智能体指数53.9→53.6-0.6%
- 分数Ring-2.6-1TInclusionAI智能指数17.3→16.6-4.0%
- 分数Ring-2.6-1TInclusionAI智能体指数12.9→10.7-17%
- 分数Trinity Large ThinkingArcee AI智能指数10.9→10.8-0.9%
- 分数Trinity Large ThinkingArcee AI智能体指数1.2→1-17%
09-17
2026年9月17日7 条变化- 新增Ling-3.0-flash-FinInclusionAI智能指数 23
- 价格HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing输入价$0.040→指标撤下
- 价格HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing输出价$0.140→指标撤下
- 价格Llama 3.3 Instruct 70BMeta输入价$0.660→$0.710+7.6%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能指数16.8→16.1-4.2%
- 分数Mistral Medium 3.1Mistral智能指数9.9→9.5-4.0%
- 分数North Mini CodeCohere智能指数12.8→9.9-23%
09-16
2026年9月16日11 条变化- 新增Qwen3.8 Max (0902)Alibaba智能指数 45.4
- 价格Claude 4 Sonnet (Non-reasoning)Anthropic输入价$3.00→指标撤下
- 价格Claude 4 Sonnet (Non-reasoning)Anthropic输出价$15.00→指标撤下
- 价格Claude 4 Sonnet (Reasoning)Anthropic输入价$3.00→指标撤下
- 价格Claude 4 Sonnet (Reasoning)Anthropic输出价$15.00→指标撤下
- 价格Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI输入价$0.500→指标撤下
- 价格Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI输出价$2.50→指标撤下
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能指数22.3→21.2-4.9%
- 分数Kimi K2.6Kimi智能指数31.3→27.5-12%
- 分数Ling 3.0 FlashInclusionAI智能指数24.9→20.6-17%
- 分数Ling-3.0-flash-VLInclusionAI智能指数24.8→25+0.8%
09-15
2026年9月15日9 条变化- 新增Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能指数 42
- 新增K2 Horizon 0.9BMBZUAI Institute of Foundation Models智能指数 3
- 新增K2 Horizon 3.7BMBZUAI Institute of Foundation Models智能指数 16.2
- 新增K2 Horizon 7BMBZUAI Institute of Foundation Models智能指数 21
- 新增K2 Horizon MoVA 36B A4BMBZUAI Institute of Foundation Models智能指数 25.7
- 价格Hy3Tencent输出价$0.550→$0.580+5.5%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输出价$2.60→$2.40-7.7%
- 分数GLM-5.2 (max)Z AI智能指数38.6→34-12%
- 分数Kimi K3 (low)Kimi智能指数34.5→30.5-12%
09-13
2026年9月13日3 条变化- 下架Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic
- 下架GPT-6 Astra (Non-reasoning)OpenAI
- 分数K2 Horizon 375B A23BMBZUAI Institute of Foundation Models智能指数30.7→30.8+0.3%
09-12
2026年9月12日6 条变化- 新增Agnes 3.0 FlashSapiens AI智能指数 35.5
- 分数Agnes 2.5 Pro AlphaSapiens AI智能指数26.8→27.8+3.7%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能指数首次评测→34.7
- 分数GPT-5.6 Luna (low)OpenAI智能指数21.8→21.5-1.4%
- 分数GPT-5.6 Luna (medium)OpenAI智能指数25.8→25.5-1.2%
- 分数K2 Horizon 375B A23BMBZUAI Institute of Foundation Models智能指数33.9→30.7-9.4%
09-11
2026年9月11日18 条变化- 新增DeepSeek V4.1 Flash (Reasoning, Max Effort)DeepSeek智能指数 39.5
- 新增Ling-3.0-flash-VLInclusionAI智能指数 24.8
- 价格GPT-6 Astra (Non-reasoning)OpenAI输入价$10.00→指标撤下
- 价格GPT-6 Astra (Non-reasoning)OpenAI输出价$50.00→指标撤下
- 价格LongCat 2.0LongCat输入价$0.750→$0.300-60%
- 价格LongCat 2.0LongCat输出价$2.95→$1.20-59%
- 分数Claude Sonnet 5 (Adaptive Reasoning, High Effort)Anthropic智能指数首次评测→32
- 分数GLM-5.1 (Reasoning)Z AI智能指数27.4→26.4-3.6%
- 分数GPT-5.6 Luna (high)OpenAI智能指数32.9→32.4-1.5%
- 分数GPT-5.6 Terra (medium)OpenAI智能指数32.8→30.4-7.3%
- 分数Granite 4.2 8BIBM智能指数12.4→11.8-4.8%
- 分数MiniCPM5-2BOpenBMB智能指数14.3→13.1-8.4%
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能指数22.3→18.8-16%
- 分数Qwen3.8 27B (low)Alibaba智能指数29.2→26.5-9.2%
- 分数Qwen3.8 27B (medium)Alibaba智能指数30.7→27.8-9.4%
- 分数Qwen3.8 27B (medium)Alibaba智能体指数首次评测→45.3
- 分数Qwen3.8-Flash-NextAlibaba智能指数42.2→39.9-5.5%
- 分数Qwen3.8-Flash-NextAlibaba智能体指数首次评测→53.9
09-09
2026年9月9日753 条变化- 新增Muse Spark 1.3 (max)Meta智能指数 48.2
- 下架Muse Spark 1.3 (max)Meta
- 价格DeepSeek V3 (Dec '24)DeepSeek输入价$0.360→$0.320-11%
- 价格DeepSeek V3 0324DeepSeek输入价$0.270→$0.840+211%
- 价格DeepSeek V3 0324DeepSeek输出价$1.12→$1.18+5.4%
- 价格Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research输入价$0.130→指标撤下
- 价格Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research输出价$0.400→指标撤下
- 价格Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research输入价$0.130→指标撤下
- 价格Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research输出价$0.400→指标撤下
- 价格Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA输入价$0.600→指标撤下
- 价格Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA输出价$1.80→指标撤下
- 价格Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA输入价$0.070→$0.090+29%
- 价格Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA输出价$0.300→$0.360+20%
- 价格Nemotron 3 Super 120B A12B (Reasoning)NVIDIA输入价$0.200→$0.190-5.0%
- 价格Nemotron 3 Super 120B A12B (Reasoning)NVIDIA输出价$0.800→$0.650-19%
- 分数A.X-K2SK Telecom智能指数26.8→22.7-15%
- 分数Agnes 2.5 Pro AlphaSapiens AI智能指数31.2→26.8-14%
- 分数Agnes 2.5 Pro AlphaSapiens AI智能体指数29.5→29.4-0.3%
- 分数Agnes 2.5 Pro BetaSapiens AI智能指数39.5→35.2-11%
- 分数Apertus 70B InstructSwiss AI Initiative智能指数1→5.1+410%
- 分数Apertus 8B InstructSwiss AI Initiative智能指数1→4.8+380%
- 分数Apodex 1.1Apodex智能指数34.9→30.4-13%
- 分数Apriel-v1.5-15B-ThinkerServiceNow智能指数15→13.8-8.0%
- 分数Apriel-v1.6-15B-ThinkerServiceNow智能指数14.3→13.4-6.3%
- 分数Arctic InstructSnowflake智能指数1→5.4+440%
- 分数Celeris-1Celeris智能指数7.6→6.3-17%
- 分数Claude 2.0Anthropic智能指数1→5.5+450%
- 分数Claude 2.1Anthropic智能指数1→5.6+460%
- 分数Claude 3 HaikuAnthropic智能指数1→5.6+460%
- 分数Claude 3 OpusAnthropic智能指数6→8.7+45%
- 分数Claude 3 SonnetAnthropic智能指数1→5.9+490%
- 分数Claude 3.5 HaikuAnthropic智能指数6.5→8.9+37%
- 分数Claude 3.5 Sonnet (June '24)Anthropic智能指数2.7→7.2+167%
- 分数Claude 3.5 Sonnet (Oct '24)Anthropic智能指数4.2→7.9+88%
- 分数Claude 3.7 Sonnet (Non-reasoning)Anthropic智能指数17.2→15.3-11%
- 分数Claude 3.7 Sonnet (Reasoning)Anthropic智能指数20.5→17.7-14%
- 分数Claude 4 Opus (Non-reasoning)Anthropic智能指数19.1→16.6-13%
- 分数Claude 4 Opus (Reasoning)Anthropic智能指数24.3→20.6-15%
- 分数Claude 4 Sonnet (Non-reasoning)Anthropic智能指数19.1→16.6-13%
- 分数Claude 4 Sonnet (Reasoning)Anthropic智能指数22.2→18.9-15%
- 分数Claude 4.1 Opus (Non-reasoning)Anthropic智能指数21.7→18.6-14%
- 分数Claude 4.1 Opus (Reasoning)Anthropic智能指数26.9→22.8-15%
- 分数Claude 4.5 Haiku (Non-reasoning)Anthropic智能指数17.4→15.4-11%
- 分数Claude 4.5 Haiku (Reasoning)Anthropic智能指数22.5→17.6-22%
- 分数Claude 4.5 Haiku (Reasoning)Anthropic智能体指数10.4→10.3-1.0%
- 分数Claude 4.5 Sonnet (Non-reasoning)Anthropic智能指数22.7→19.3-15%
- 分数Claude 4.5 Sonnet (Reasoning)Anthropic智能指数27.3→21.2-22%
- 分数Claude 4.5 Sonnet (Reasoning)Anthropic智能体指数17.6→17.5-0.6%
- 分数Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic智能指数53.2→49.7-6.6%
- 分数Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic智能体指数51.2→51-0.4%
- 分数Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic智能指数54.5→51.2-6.1%
- 分数Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic智能体指数53.5→53.3-0.4%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic智能指数50.9→47-7.7%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic智能体指数48→47.8-0.4%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic智能指数56.8→53.4-6.0%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic智能体指数58.2→58-0.3%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic智能指数52.8→49.1-7.0%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic智能体指数50.9→50.6-0.6%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic智能指数56.2→53.2-5.3%
- 分数Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic智能体指数57.4→57.1-0.5%
- 分数Claude InstantAnthropic智能指数1→5+400%
- 分数Claude Opus 4.5 (Non-reasoning)Anthropic智能指数27.8→23.7-15%
- 分数Claude Opus 4.5 (Reasoning)Anthropic智能指数33.6→29.1-13%
- 分数Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能指数36.4→31.9-12%
- 分数Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic智能指数30.8→26.4-14%
- 分数Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic智能指数44.3→40.7-8.1%
- 分数Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic智能体指数39.7→39.5-0.5%
- 分数Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic智能指数35.4→30.9-13%
- 分数Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能指数47.8→42-12%
- 分数Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能体指数42.8→42.6-0.5%
- 分数Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能指数52→48.2-7.3%
- 分数Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能体指数52.9→52.7-0.4%
- 分数Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能指数43.8→39.8-9.1%
- 分数Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能体指数37.7→37.5-0.5%
- 分数Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数54.1→50.7-6.3%
- 分数Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能体指数56.4→56.2-0.4%
- 分数Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能指数49.5→45.1-8.9%
- 分数Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能体指数47.2→47-0.4%
- 分数Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能指数53.4→49.7-6.9%
- 分数Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能体指数55.7→55.5-0.4%
- 分数Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能指数38.5→30.5-21%
- 分数Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能体指数33.3→33.1-0.6%
- 分数Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic智能指数29→24.7-15%
- 分数Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic智能指数27.4→23.3-15%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Low Effort)Anthropic智能指数首次评测→24.7
- 分数Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数45.1→38.4-15%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic智能体指数44.5→44.3-0.4%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)Anthropic智能指数首次评测→28.4
- 分数Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic智能指数33.5→28.9-14%
- 分数Command ACohere智能指数2.1→7+233%
- 分数Command A+Cohere智能指数15.6→13.9-11%
- 分数Command A+Cohere智能体指数3.7→3.6-2.7%
- 分数Command-R (Mar '24)Cohere智能指数1→5+400%
- 分数Command-R+ (Apr '24)Cohere智能指数1→5.3+430%
- 分数DBRX InstructDatabricks智能指数1→5.3+430%
- 分数DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research智能指数1→5.1+410%
- 分数DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research智能指数1→6.1+510%
- 分数DeepSeek Coder V2 Lite InstructDeepSeek智能指数1→5.3+430%
- 分数DeepSeek LLM 67B Chat (V1)DeepSeek智能指数1→5.3+430%
- 分数DeepSeek R1 (Jan '25)DeepSeek智能指数12→11.4-5.0%
- 分数DeepSeek R1 0528 (May '25)DeepSeek智能指数13.9→13.1-5.8%
- 分数DeepSeek R1 0528 Qwen3 8BDeepSeek智能指数4.7→8.1+72%
- 分数DeepSeek R1 Distill Llama 70BDeepSeek智能指数4.2→7.9+88%
- 分数DeepSeek R1 Distill Llama 8BDeepSeek智能指数1→6.5+550%
- 分数DeepSeek R1 Distill Qwen 1.5BDeepSeek智能指数1→5.5+450%
- 分数DeepSeek R1 Distill Qwen 14BDeepSeek智能指数4.1→7.8+90%
- 分数DeepSeek R1 Distill Qwen 32BDeepSeek智能指数5.3→8.4+58%
- 分数DeepSeek V3 (Dec '24)DeepSeek智能指数10.4→8.5-18%
- 分数DeepSeek V3 0324DeepSeek智能指数9.2→9.7+5.4%
- 分数DeepSeek V3.1 (Non-reasoning)DeepSeek智能指数14.8→13.7-7.4%
- 分数DeepSeek V3.1 (Reasoning)DeepSeek智能指数14.5→13.5-6.9%
- 分数DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek智能指数15.2→13.9-8.6%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能指数23.5→15.4-34%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能体指数9→8.9-1.1%
- 分数DeepSeek V3.2 (Non-reasoning)DeepSeek智能指数18.3→16-13%
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能指数25.3→21.5-15%
- 分数DeepSeek V3.2 Exp (Non-reasoning)DeepSeek智能指数15.1→13.9-7.9%
- 分数DeepSeek V3.2 Exp (Reasoning)DeepSeek智能指数19→16.6-13%
- 分数DeepSeek V3.2 SpecialeDeepSeek智能指数16→14.5-9.4%
- 分数DeepSeek V4 Flash 0420 (Non-reasoning)DeepSeek智能指数22.1→18.9-14%
- 分数DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek智能指数30.2→24.8-18%
- 分数DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek智能体指数28.1→27.9-0.7%
- 分数DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek智能指数30.7→24.6-20%
- 分数DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek智能体指数23.8→23.7-0.4%
- 分数DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能指数40.8→34.5-15%
- 分数DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能体指数41.9→41.7-0.5%
- 分数DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek智能指数40.7→35-14%
- 分数DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek智能体指数48.2→48.1-0.2%
- 分数DeepSeek V4 Pro 0424 (Non-reasoning)DeepSeek智能指数24.5→20.8-15%
- 分数DeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeek智能指数34.6→30.1-13%
- 分数DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek智能指数34.5→30.9-10%
- 分数DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek智能体指数27.9→27.7-0.7%
- 分数DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek智能指数42.1→36.3-14%
- 分数DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek智能体指数42.5→42.3-0.5%
- 分数DeepSeek-Coder-V2DeepSeek智能指数1→6+500%
- 分数DeepSeek-V2-ChatDeepSeek智能指数1→5.5+450%
- 分数DeepSeek-V2.5DeepSeek智能指数1.1→6.6+500%
- 分数DeepSeek-V2.5 (Dec '24)DeepSeek智能指数1.2→6.6+450%
- 分数Devstral 2Mistral智能指数12.7→9.4-26%
- 分数Devstral MediumMistral智能指数6.6→9+36%
- 分数Devstral Small (Jul '25)Mistral智能指数3.6→7.6+111%
- 分数Devstral Small (May '25)Mistral智能指数6→8.7+45%
- 分数Devstral Small 2Mistral智能指数11.7→8.4-28%
- 分数DiffusionGemma 26B A4BGoogle智能指数7.6→9.5+25%
- 分数Doubao Seed CodeByteDance Seed智能指数19.5→16.9-13%
- 分数ERNIE 4.5 300B A47BBaidu智能指数3.4→7.5+121%
- 分数ERNIE 5.0 Thinking PreviewBaidu智能指数15.7→14.3-8.9%
- 分数Exaone 4.0 1.2B (Non-reasoning)LG AI Research智能指数1→5.2+420%
- 分数Exaone 4.0 1.2B (Reasoning)LG AI Research智能指数1→5.3+430%
- 分数EXAONE 4.0 32B (Non-reasoning)LG AI Research智能指数1→6.3+530%
- 分数EXAONE 4.0 32B (Reasoning)LG AI Research智能指数4.9→8.2+67%
- 分数EXAONE 4.5 33BLG AI Research智能指数14→13.2-5.7%
- 分数Falcon-H1R-7BTII UAE智能指数4.1→7.8+90%
- 分数G9v3-39A5BAI9Stars智能指数25.7→21.8-15%
- 分数G9v3-3BAI9Stars智能指数10.1→10.8+6.9%
- 分数Gemini 1.0 ProGoogle智能指数1→5.3+430%
- 分数Gemini 1.0 UltraGoogle智能指数1→5.8+480%
- 分数Gemini 1.5 Flash (May '24)Google智能指数1→5.9+490%
- 分数Gemini 1.5 Flash (Sep '24)Google智能指数2.4→7.1+196%
- 分数Gemini 1.5 Flash-8BGoogle智能指数1→6.2+520%
- 分数Gemini 1.5 Pro (May '24)Google智能指数1→6.4+540%
- 分数Gemini 1.5 Pro (Sep '24)Google智能指数4.3→7.9+84%
- 分数Gemini 2.0 Flash (experimental)Google智能指数5→8.2+64%
- 分数Gemini 2.0 Flash (Feb '25)Google智能指数6.5→8.9+37%
- 分数Gemini 2.0 Flash Thinking Experimental (Dec '24)Google智能指数1.1→6.6+500%
- 分数Gemini 2.0 Flash Thinking Experimental (Jan '25)Google智能指数7.4→9.4+27%
- 分数Gemini 2.0 Flash-Lite (Feb '25)Google智能指数3.2→7.4+131%
- 分数Gemini 2.0 Flash-Lite (Preview)Google智能指数3→7.3+143%
- 分数Gemini 2.0 Pro Experimental (Feb '25)Google智能指数6.1→8.7+43%
- 分数Gemini 2.5 Flash (Non-reasoning)Google智能指数8.2→9.9+21%
- 分数Gemini 2.5 Flash (Reasoning)Google智能指数13.9→13.1-5.8%
- 分数Gemini 2.5 Flash Preview (Non-reasoning)Google智能指数5.9→8.7+47%
- 分数Gemini 2.5 Flash Preview (Reasoning)Google智能指数11.5→11.7+1.7%
- 分数Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google智能指数12.7→12.4-2.4%
- 分数Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google智能指数17.4→15.5-11%
- 分数Gemini 2.5 Flash-Lite (Non-reasoning)Google智能指数1.4→6.7+379%
- 分数Gemini 2.5 Flash-Lite (Reasoning)Google智能指数5.6→8.5+52%
- 分数Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google智能指数7.3→9.3+27%
- 分数Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google智能指数9.2→10.4+13%
- 分数Gemini 2.5 ProGoogle智能指数19→16.7-12%
- 分数Gemini 2.5 ProGoogle智能体指数3.6→3.5-2.8%
- 分数Gemini 2.5 Pro Preview (Mar' 25)Google智能指数16.7→15-10%
- 分数Gemini 2.5 Pro Preview (May' 25)Google智能指数16.1→14.5-9.9%
- 分数Gemini 3 Flash Preview (Non-reasoning)Google智能指数20.8→17.9-14%
- 分数Gemini 3 Flash Preview (Reasoning)Google智能指数30.7→26.3-14%
- 分数Gemini 3 Pro Preview (high)Google智能指数32.4→28-14%
- 分数Gemini 3 Pro Preview (low)Google智能指数26.3→22.3-15%
- 分数Gemini 3.1 Flash-LiteGoogle智能指数19.3→16-17%
- 分数Gemini 3.1 Flash-LiteGoogle智能体指数3.3→3.2-3.0%
- 分数Gemini 3.1 Pro PreviewGoogle智能指数36.7→30.4-17%
- 分数Gemini 3.1 Pro PreviewGoogle智能体指数10.4→10.3-1.0%
- 分数Gemini 3.5 Flash (high)Google智能指数39.7→33-17%
- 分数Gemini 3.5 Flash (high)Google智能体指数27.4→27.3-0.4%
- 分数Gemini 3.5 Flash (medium)Google智能指数38→33.6-12%
- 分数Gemini 3.5 Flash (minimal)Google智能指数28→23.8-15%
- 分数Gemini 3.5 Flash-LiteGoogle智能指数27.6→22.7-18%
- 分数Gemini 3.5 Flash-LiteGoogle智能体指数16.1→15.9-1.2%
- 分数Gemini 3.6 Flash (high)Google智能指数40.3→34.3-15%
- 分数Gemini 3.6 Flash (high)Google智能体指数30.3→30.2-0.3%
- 分数Gemini 3.7 Flash (high)Google智能指数45.2→39.4-13%
- 分数Gemini 3.7 Flash (high)Google智能体指数36.6→36.4-0.5%
- 分数Gemini 3.7 Flash (low)Google智能指数41.1→36.9-10%
- 分数Gemini 3.7 Flash (medium)Google智能指数43.4→39.6-8.8%
- 分数Gemini 3.8 Flash (high)Google智能指数47.1→41.2-13%
- 分数Gemini 3.8 Flash (high)Google智能体指数41.2→41.1-0.2%
- 分数Gemini 3.8 Flash (low)Google智能指数41→33.8-18%
- 分数Gemini 3.8 Flash (low)Google智能体指数33.8→33.7-0.3%
- 分数Gemini 3.8 Flash (medium)Google智能指数46.8→40-15%
- 分数Gemini 3.8 Flash (medium)Google智能体指数首次评测→40.5
- 分数Gemma 3 12B InstructGoogle智能指数1→3.8+280%
- 分数Gemma 3 1B InstructGoogle智能指数1→4.8+380%
- 分数Gemma 3 270MGoogle智能指数1→5.1+410%
- 分数Gemma 3 27B InstructGoogle智能指数2→4.9+145%
- 分数Gemma 3 4B InstructGoogle智能指数1→4.8+380%
- 分数Gemma 3n E2B InstructGoogle智能指数1→4.8+380%
- 分数Gemma 3n E4B InstructGoogle智能指数1→4.8+380%
- 分数Gemma 3n E4B Instruct Preview (May '25)Google智能指数1→5.8+480%
- 分数Gemma 4 12B (Non-reasoning)Google智能指数7.3→9.4+29%
- 分数Gemma 4 12B (Reasoning)Google智能指数15.6→14.2-9.0%
- 分数Gemma 4 26B A4B (Non-reasoning)Google智能指数13.9→13.1-5.8%
- 分数Gemma 4 26B A4B (Reasoning)Google智能指数19.2→16.7-13%
- 分数Gemma 4 31B (Non-reasoning)Google智能指数15.2→13.9-8.6%
- 分数Gemma 4 31B (Reasoning)Google智能指数22.2→15.4-31%
- 分数Gemma 4 31B (Reasoning)Google智能体指数6.8→6.7-1.5%
- 分数Gemma 4 E2B (Non-reasoning)Google智能指数1→6.5+550%
- 分数Gemma 4 E2B (Reasoning)Google智能指数4→7.8+95%
- 分数Gemma 4 E4B (Non-reasoning)Google智能指数3.3→7.5+127%
- 分数Gemma 4 E4B (Reasoning)Google智能指数6.4→8.9+39%
- 分数GLM 5V Turbo (Reasoning)Z AI智能指数27.6→23.5-15%
- 分数GLM-4.5 (Reasoning)Z AI智能指数13.3→12.8-3.8%
- 分数GLM-4.5-AirZ AI智能指数10.5→11.1+5.7%
- 分数GLM-4.5V (Non-reasoning)Z AI智能指数1.4→6.7+379%
- 分数GLM-4.5V (Reasoning)Z AI智能指数3.5→7.6+117%
- 分数GLM-4.6 (Non-reasoning)Z AI智能指数16.7→14.9-11%
- 分数GLM-4.6 (Reasoning)Z AI智能指数21.7→18.5-15%
- 分数GLM-4.6V (Non-reasoning)Z AI智能指数5.2→8.4+62%
- 分数GLM-4.6V (Reasoning)Z AI智能指数10.7→11.2+4.7%
- 分数GLM-4.7 (Non-reasoning)Z AI智能指数20.1→17.4-13%
- 分数GLM-4.7 (Reasoning)Z AI智能指数26.2→22.2-15%
- 分数GLM-4.7-Flash (Non-reasoning)Z AI智能指数9.6→10.6+10%
- 分数GLM-4.7-Flash (Reasoning)Z AI智能指数16.6→14.9-10%
- 分数GLM-5 (Non-reasoning)Z AI智能指数25.6→21.8-15%
- 分数GLM-5 (Reasoning)Z AI智能指数32.4→27.9-14%
- 分数GLM-5-TurboZ AI智能指数31→26.6-14%
- 分数GLM-5.1 (Non-reasoning)Z AI智能指数28.5→24.2-15%
- 分数GLM-5.1 (Reasoning)Z AI智能指数31.9→27.4-14%
- 分数GLM-5.1 (Reasoning)Z AI智能体指数25.4→25.2-0.8%
- 分数GLM-5.2 (max)Z AI智能指数42.1→38.6-8.3%
- 分数GLM-5.2 (max)Z AI智能体指数39.7→39.4-0.8%
- 分数GLM-5.2 (Non-reasoning)Z AI智能指数26.5→22.4-15%
- 分数GLM-5.3 (max)Z AI智能指数48.6→44.9-7.6%
- 分数GLM-5.3 (max)Z AI智能体指数53.6→53.4-0.4%
- 分数GLM 5.3 FlashZ AI智能指数46.2→41.9-9.3%
- 分数GLM 5.3 FlashZ AI智能体指数51.5→51.2-0.6%
- 分数GPT-3.5 TurboOpenAI智能指数1→5.5+450%
- 分数GPT-4OpenAI智能指数1.5→6.7+347%
- 分数GPT-4 TurboOpenAI智能指数2.3→7+204%
- 分数GPT-4.1OpenAI智能指数13.2→12.7-3.8%
- 分数GPT-4.1 miniOpenAI智能指数8.8→10.2+16%
- 分数GPT-4.1 nanoOpenAI智能指数4.1→7.8+90%
- 分数GPT-4.5 (Preview)OpenAI智能指数7.7→9.6+25%
- 分数GPT-4o (Aug '24)OpenAI智能指数3.9→7.7+97%
- 分数GPT-4o (ChatGPT)OpenAI智能指数2.6→7.2+177%
- 分数GPT-4o (March 2025, chatgpt-4o-latest)OpenAI智能指数6.5→9+38%
- 分数GPT-4o (May '24)OpenAI智能指数3→7.3+143%
- 分数GPT-4o (Nov '24)OpenAI智能指数5.4→8.4+56%
- 分数GPT-4o miniOpenAI智能指数1.4→6.7+379%
- 分数GPT-5 (ChatGPT)OpenAI智能指数9.4→10.4+11%
- 分数GPT-5 (high)OpenAI智能指数27.1→23-15%
- 分数GPT-5 (low)OpenAI智能指数24.4→20.8-15%
- 分数GPT-5 (medium)OpenAI智能指数26.9→22.9-15%
- 分数GPT-5 (minimal)OpenAI智能指数11.1→11.4+2.7%
- 分数GPT-5 Codex (high)OpenAI智能指数29.2→24.9-15%
- 分数GPT-5 mini (high)OpenAI智能指数18.4→17.4-5.4%
- 分数GPT-5 mini (high)OpenAI智能体指数9→8.9-1.1%
- 分数GPT-5 mini (medium)OpenAI智能指数24.2→20.6-15%
- 分数GPT-5 mini (minimal)OpenAI智能指数8.4→9.9+18%
- 分数GPT-5 nano (high)OpenAI智能指数13.7→13-5.1%
- 分数GPT-5 nano (medium)OpenAI智能指数12.9→12.5-3.1%
- 分数GPT-5 nano (minimal)OpenAI智能指数2.4→7.1+196%
- 分数GPT-5.1 (high)OpenAI智能指数29.1→24.7-15%
- 分数GPT-5.1 (Non-reasoning)OpenAI智能指数14.2→13.3-6.3%
- 分数GPT-5.1 Codex (high)OpenAI智能指数27.9→23.7-15%
- 分数GPT-5.1 Codex mini (high)OpenAI智能指数24→20.4-15%
- 分数GPT-5.2 (medium)OpenAI智能指数30.9→26.5-14%
- 分数GPT-5.2 (Non-reasoning)OpenAI智能指数19.6→17-13%
- 分数GPT-5.2 (xhigh)OpenAI智能指数35→30.4-13%
- 分数GPT-5.2 Codex (xhigh)OpenAI智能指数33→28.5-14%
- 分数GPT-5.3 Codex (xhigh)OpenAI智能指数36.9→32.5-12%
- 分数GPT-5.4 (low)OpenAI智能指数32.1→27.6-14%
- 分数GPT-5.4 (Non-reasoning)OpenAI智能指数21.1→18.2-14%
- 分数GPT-5.4 (xhigh)OpenAI智能指数42.8→39-8.9%
- 分数GPT-5.4 mini (medium)OpenAI智能指数23.2→19.7-15%
- 分数GPT-5.4 mini (Non-Reasoning)OpenAI智能指数10.6→11.1+4.7%
- 分数GPT-5.4 mini (xhigh)OpenAI智能指数31.9→24.6-23%
- 分数GPT-5.4 mini (xhigh)OpenAI智能体指数19.8→19.7-0.5%
- 分数GPT-5.4 nano (medium)OpenAI智能指数23.5→20-15%
- 分数GPT-5.4 nano (Non-Reasoning)OpenAI智能指数11.5→11.7+1.7%
- 分数GPT-5.4 nano (xhigh)OpenAI智能指数30.9→21.2-31%
- 分数GPT-5.4 nano (xhigh)OpenAI智能体指数17.9→17.7-1.1%
- 分数GPT-5.5 (high)OpenAI智能指数44.1→37.3-15%
- 分数GPT-5.5 (high)OpenAI智能体指数35.6→35.4-0.6%
- 分数GPT-5.5 (low)OpenAI智能指数35.3→30.7-13%
- 分数GPT-5.5 (medium)OpenAI智能指数41.2→34.2-17%
- 分数GPT-5.5 (medium)OpenAI智能体指数30.7→30.6-0.3%
- 分数GPT-5.5 (Non-reasoning)OpenAI智能指数27.3→23.2-15%
- 分数GPT-5.5 (xhigh)OpenAI智能指数45.6→38.6-15%
- 分数GPT-5.5 (xhigh)OpenAI智能体指数37.5→37.3-0.5%
- 分数GPT-5.5 Instant (June 2026)OpenAI智能指数21.6→26.8+24%
- 分数GPT-5.5 Instant (June 2026)OpenAI智能体指数19.4→19.2-1.0%
- 分数GPT-5.5 Instant (May 2026)OpenAI智能指数26.7→22.7-15%
- 分数GPT-5.6 Luna (high)OpenAI智能指数37.4→32.9-12%
- 分数GPT-5.6 Luna (high)OpenAI智能体指数35.8→35.6-0.6%
- 分数GPT-5.6 Luna (low)OpenAI智能指数25.8→21.8-16%
- 分数GPT-5.6 Luna (low)OpenAI智能体指数18→17.9-0.6%
- 分数GPT-5.6 Luna (max)OpenAI智能指数43.4→37.5-14%
- 分数GPT-5.6 Luna (max)OpenAI智能体指数42.9→42.7-0.5%
- 分数GPT-5.6 Luna (medium)OpenAI智能指数30.2→25.8-15%
- 分数GPT-5.6 Luna (medium)OpenAI智能体指数25.5→25.4-0.4%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能指数19.3→16.8-13%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能体指数15.3→15.2-0.7%
- 分数GPT-5.6 Luna (xhigh)OpenAI智能指数41.6→34.8-16%
- 分数GPT-5.6 Luna (xhigh)OpenAI智能体指数39.7→39.5-0.5%
- 分数GPT-5.6 Sol (high)OpenAI智能指数48.3→42.5-12%
- 分数GPT-5.6 Sol (high)OpenAI智能体指数44.9→44.7-0.4%
- 分数GPT-5.6 Sol (low)OpenAI智能指数40.8→33.8-17%
- 分数GPT-5.6 Sol (low)OpenAI智能体指数32.9→32.6-0.9%
- 分数GPT-5.6 Sol (max)OpenAI智能指数51.3→47.1-8.2%
- 分数GPT-5.6 Sol (max)OpenAI智能体指数50.7→50.5-0.4%
- 分数GPT-5.6 Sol (medium)OpenAI智能指数46→39.5-14%
- 分数GPT-5.6 Sol (medium)OpenAI智能体指数40.8→40.5-0.7%
- 分数GPT-5.6 Sol (Non-reasoning)OpenAI智能指数32.9→28.3-14%
- 分数GPT-5.6 Sol (Non-reasoning)OpenAI智能体指数29.5→29.4-0.3%
- 分数GPT-5.6 Sol (xhigh)OpenAI智能指数49.8→44.1-11%
- 分数GPT-5.6 Sol (xhigh)OpenAI智能体指数48.1→47.8-0.6%
- 分数GPT-5.6 Terra (high)OpenAI智能指数41.3→34.5-16%
- 分数GPT-5.6 Terra (high)OpenAI智能体指数37.8→37.6-0.5%
- 分数GPT-5.6 Terra (low)OpenAI智能指数32.4→27.9-14%
- 分数GPT-5.6 Terra (max)OpenAI智能指数46.8→42.3-9.6%
- 分数GPT-5.6 Terra (max)OpenAI智能体指数43.9→43.7-0.5%
- 分数GPT-5.6 Terra (medium)OpenAI智能指数37.2→32.8-12%
- 分数GPT-5.6 Terra (medium)OpenAI智能体指数31.6→31.4-0.6%
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能指数26.3→22.3-15%
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能体指数25.1→24.9-0.8%
- 分数GPT-5.6 Terra (xhigh)OpenAI智能指数44.4→38.2-14%
- 分数GPT-5.6 Terra (xhigh)OpenAI智能体指数42.4→42.2-0.5%
- 分数GPT-6 Astra (high)OpenAI智能指数53.4→51-4.5%
- 分数GPT-6 Astra (high)OpenAI智能体指数48.9→48.7-0.4%
- 分数GPT-6 Astra (low)OpenAI智能指数49.3→46-6.7%
- 分数GPT-6 Astra (low)OpenAI智能体指数39.7→39.5-0.5%
- 分数GPT-6 Astra (max)OpenAI智能指数54.7→52.8-3.5%
- 分数GPT-6 Astra (max)OpenAI智能体指数51.6→51.5-0.2%
- 分数GPT-6 Astra (medium)OpenAI智能指数52.2→49.7-4.8%
- 分数GPT-6 Astra (medium)OpenAI智能体指数46.5→46.3-0.4%
- 分数GPT-6 Astra (Non-reasoning)OpenAI智能指数47.8→45.2-5.4%
- 分数GPT-6 Astra (Non-reasoning)OpenAI智能体指数48→47.8-0.4%
- 分数GPT-6 Astra (xhigh)OpenAI智能指数54.3→52.5-3.3%
- 分数GPT-6 Astra (xhigh)OpenAI智能体指数50.8→50.6-0.4%
- 分数gpt-oss-120b (high)OpenAI智能指数15.6→12.3-21%
- 分数gpt-oss-120b (high)OpenAI智能体指数6.3→6.2-1.6%
- 分数gpt-oss-120b (low)OpenAI智能指数8.9→10.2+15%
- 分数gpt-oss-20b (high)OpenAI智能指数10.8→9-17%
- 分数gpt-oss-20b (low)OpenAI智能指数8.4→10+19%
- 分数Granite 3.3 8B (Non-reasoning)IBM智能指数1→4.9+390%
- 分数Granite 4.0 1BIBM智能指数1→5+400%
- 分数Granite 4.0 350MIBM智能指数1→4.8+380%
- 分数Granite 4.0 H 1BIBM智能指数1→5.2+420%
- 分数Granite 4.0 H 350MIBM智能指数1→4.8+380%
- 分数Granite 4.0 H SmallIBM智能指数1→6+500%
- 分数Granite 4.0 MicroIBM智能指数1→5.1+410%
- 分数Granite 4.1 30BIBM智能指数3.2→7.4+131%
- 分数Granite 4.1 3BIBM智能指数1→5.9+490%
- 分数Granite 4.1 8BIBM智能指数1.1→6.6+500%
- 分数Granite 4.2 30BIBM智能指数16.5→14.8-10%
- 分数Granite 4.2 3BIBM智能指数10.7→9.1-15%
- 分数Granite 4.2 8BIBM智能指数13.9→12.4-11%
- 分数Granite 4.2 8BIBM智能体指数3.8→3.7-2.6%
- 分数Grok 2 (Dec '24)SpaceXAI智能指数2.4→7.1+196%
- 分数Grok 3SpaceXAI智能指数12.3→12.1-1.6%
- 分数Grok 3 mini Reasoning (high)SpaceXAI智能指数16.2→14.6-9.9%
- 分数Grok 3 Reasoning BetaSpaceXAI智能指数9.2→10.4+13%
- 分数Grok 4SpaceXAI智能指数26.5→22.5-15%
- 分数Grok 4 Fast (Non-reasoning)SpaceXAI智能指数10.5→11.1+5.7%
- 分数Grok 4 Fast (Reasoning)SpaceXAI智能指数20.8→17.9-14%
- 分数Grok 4.1 Fast (Non-reasoning)SpaceXAI智能指数10.9→11.3+3.7%
- 分数Grok 4.1 Fast (Reasoning)SpaceXAI智能指数23.9→20.4-15%
- 分数Grok 4.20 0309 (Non-reasoning)SpaceXAI智能指数16.2→14.6-9.9%
- 分数Grok 4.20 0309 (Reasoning)SpaceXAI智能指数29.5→25.2-15%
- 分数Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI智能指数15.6→14.2-9.0%
- 分数Grok 4.20 0309 v2 (Reasoning)SpaceXAI智能指数30→25.7-14%
- 分数Grok 4.3 (high)SpaceXAI智能指数29.9→25.4-15%
- 分数Grok 4.3 (high)SpaceXAI智能体指数17.3→17.2-0.6%
- 分数Grok 4.3 (low)SpaceXAI智能指数28.5→24.3-15%
- 分数Grok 4.3 (medium)SpaceXAI智能指数29.1→24.8-15%
- 分数Grok 4.3 (Non-reasoning)SpaceXAI智能指数17.5→14.5-17%
- 分数Grok 4.3 (Non-reasoning)SpaceXAI智能体指数首次评测→15.8
- 分数Grok 4.5 (high)SpaceXAI智能指数45.5→39.1-14%
- 分数Grok 4.5 (high)SpaceXAI智能体指数42.4→42.1-0.7%
- 分数Grok 4.6 (high)SpaceXAI智能指数50.6→44.4-12%
- 分数Grok 4.6 (high)SpaceXAI智能体指数53.6→53.4-0.4%
- 分数Grok 4.6 (low)SpaceXAI智能指数41.6→35.4-15%
- 分数Grok 4.6 (low)SpaceXAI智能体指数42.4→42.2-0.5%
- 分数Grok 4.6 (medium)SpaceXAI智能指数48.8→43-12%
- 分数Grok 4.6 (medium)SpaceXAI智能体指数51.4→51.2-0.4%
- 分数Grok 4.6 (xhigh)SpaceXAI智能指数49.9→44.3-11%
- 分数Grok 4.6 (xhigh)SpaceXAI智能体指数53→52.7-0.6%
- 分数Grok BetaSpaceXAI智能指数1.9→6.9+263%
- 分数Grok Build 0.1 0616SpaceXAI智能指数31.7→27.2-14%
- 分数Grok Code Fast 1SpaceXAI智能指数15.4→14.1-8.4%
- 分数Grok-1SpaceXAI智能指数1→6.3+530%
- 分数Hermes 3 - Llama-3.1 70BNous Research智能指数1→6+500%
- 分数Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research智能指数3.2→7.4+131%
- 分数Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research智能指数3.4→7.5+121%
- 分数Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research智能指数1.3→6.7+415%
- 分数Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research智能指数4.3→7.9+84%
- 分数Hy3Tencent智能指数32.4→25.8-20%
- 分数Hy3Tencent智能体指数25.7→25.6-0.4%
- 分数Hy3-preview (Non-reasoning)Tencent智能指数19.6→17-13%
- 分数Hy3-preview (Reasoning)Tencent智能指数26.8→22.7-15%
- 分数HyperCLOVA X SEED Think (32B)Naver智能指数11→11.4+3.6%
- 分数Inkling (xhigh)Thinking Machines智能指数32.2→25.5-21%
- 分数Inkling (xhigh)Thinking Machines智能体指数24.4→24.3-0.4%
- 分数Inkling SmallThinking Machines智能指数32.2→26.1-19%
- 分数Inkling SmallThinking Machines智能体指数25.1→25-0.4%
- 分数INTELLECT-3 (based on GLM-4.5-Air)Prime Intellect智能指数9.6→10.6+10%
- 分数Jamba 1.5 LargeAI21 Labs智能指数1→6+500%
- 分数Jamba 1.5 MiniAI21 Labs智能指数1→5.2+420%
- 分数Jamba 1.6 LargeAI21 Labs智能指数1→6+500%
- 分数Jamba 1.6 MiniAI21 Labs智能指数1→5.2+420%
- 分数Jamba 1.7 LargeAI21 Labs智能指数1→6.1+510%
- 分数Jamba 1.7 MiniAI21 Labs智能指数1→5.2+420%
- 分数Jamba Reasoning 3BAI21 Labs智能指数1→5.7+470%
- 分数JT-35B-FlashChina Mobile智能指数21.8→18.7-14%
- 分数JT-4.1 Flash 236B A21BChina Mobile智能指数31.8→27.3-14%
- 分数JT-MINIChina Mobile智能指数12.4→12.2-1.6%
- 分数K-EXAONE (Non-reasoning)LG AI Research智能指数10.7→11.2+4.7%
- 分数K-EXAONE (Reasoning)LG AI Research智能指数15.8→14.4-8.9%
- 分数K-EXAONE 2.0 0803LG AI Research智能指数23.1→19.7-15%
- 分数K2 Horizon 375B A23BMBZUAI Institute of Foundation Models智能指数37.8→33.9-10%
- 分数K2 Horizon 375B A23BMBZUAI Institute of Foundation Models智能体指数41.2→40.9-0.7%
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能指数11.2→11.5+2.7%
- 分数K2-V2 (high)MBZUAI Institute of Foundation Models智能指数8.3→9.9+19%
- 分数K2-V2 (low)MBZUAI Institute of Foundation Models智能指数2.9→7.3+152%
- 分数K2-V2 (medium)MBZUAI Institute of Foundation Models智能指数6.6→9+36%
- 分数KAT Coder Pro V2KwaiKAT智能指数25.6→21.7-15%
- 分数KAT-Coder-Pro V1KwaiKAT智能指数21.7→18.6-14%
- 分数Kimi K2Kimi智能指数13.3→12.7-4.5%
- 分数Kimi K2 0905Kimi智能指数17.2→15.3-11%
- 分数Kimi K2 ThinkingKimi智能指数25.9→22-15%
- 分数Kimi K2.5 (Non-reasoning)Kimi智能指数22.8→19.4-15%
- 分数Kimi K2.5 (Reasoning)Kimi智能指数27.6→23.5-15%
- 分数Kimi K2.6Kimi智能指数35.8→31.3-13%
- 分数Kimi K2.6Kimi智能体指数22.2→22.1-0.5%
- 分数Kimi K2.6 (Non-reasoning)Kimi智能指数27.7→23.6-15%
- 分数Kimi K2.7 CodeKimi智能指数32.7→26.3-20%
- 分数Kimi K2.7 CodeKimi智能体指数22.6→22.5-0.4%
- 分数Kimi K3 (low)Kimi智能指数37.1→34.5-7.0%
- 分数Kimi K3 (low)Kimi智能体指数30.8→30.7-0.3%
- 分数Kimi K3 (max)Kimi智能指数50.2→43.8-13%
- 分数Kimi K3 (max)Kimi智能体指数50.9→50.6-0.6%
- 分数Kimi Linear 48B A3B InstructKimi智能指数2.9→7.3+152%
- 分数LFM 40BLiquid AI智能指数1→5.4+440%
- 分数LFM2 1.2BLiquid AI智能指数1→4.8+380%
- 分数LFM2 2.6BLiquid AI智能指数1→5.2+420%
- 分数LFM2 24B A2BLiquid AI智能指数1→5.9+490%
- 分数LFM2 8B A1BLiquid AI智能指数1→4.9+390%
- 分数LFM2.5-1.2B-InstructLiquid AI智能指数1→5.2+420%
- 分数LFM2.5-1.2B-ThinkingLiquid AI智能指数1→5.2+420%
- 分数LFM2.5-2.6BLiquid AI智能指数5.3→8.4+58%
- 分数LFM2.5-8B-A1BLiquid AI智能指数2.7→7.2+167%
- 分数LFM2.5-VL-1.6BLiquid AI智能指数1→4.8+380%
- 分数Ling 2.6 FlashInclusionAI智能指数7.9→9.7+23%
- 分数Ling 3.0 FlashInclusionAI智能指数27.4→24.9-9.1%
- 分数Ling 3.0 FlashInclusionAI智能体指数21.1→21-0.5%
- 分数Ling 3.0 TinyInclusionAI智能指数15.7→11.9-24%
- 分数Ling-1TInclusionAI智能指数6.9→9.2+33%
- 分数Ling-2.6-1TInclusionAI智能指数19.6→17-13%
- 分数Ling-flash-2.0InclusionAI智能指数4.1→7.8+90%
- 分数Ling-mini-2.0InclusionAI智能指数1→5.5+450%
- 分数Llama 2 Chat 13BMeta智能指数1→5.3+430%
- 分数Llama 2 Chat 70BMeta智能指数1→5.3+430%
- 分数Llama 2 Chat 7BMeta智能指数1→5.7+470%
- 分数Llama 3 Instruct 70BMeta智能指数1→5.5+450%
- 分数Llama 3 Instruct 8BMeta智能指数1→4.8+380%
- 分数Llama 3.1 Instruct 405BMeta智能指数2.9→7.3+152%
- 分数Llama 3.1 Instruct 70BMeta智能指数1.2→6.6+450%
- 分数Llama 3.1 Instruct 8BMeta智能指数2→6.9+245%
- 分数Llama 3.1 Nemotron Instruct 70BNVIDIA智能指数2.1→6.9+229%
- 分数Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIA智能指数2.9→7.3+152%
- 分数Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA智能指数3.4→7.5+121%
- 分数Llama 3.1 Tulu3 405BAllen Institute for AI智能指数2.7→7.2+167%
- 分数Llama 3.2 Instruct 11B (Vision)Meta智能指数1→5.4+440%
- 分数Llama 3.2 Instruct 1BMeta智能指数1→4.8+380%
- 分数Llama 3.2 Instruct 3BMeta智能指数1→5.7+470%
- 分数Llama 3.2 Instruct 90B (Vision)Meta智能指数1→6.4+540%
- 分数Llama 3.3 Instruct 70BMeta智能指数3.7→7.7+108%
- 分数Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA智能指数2.8→7.3+161%
- 分数Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIA智能指数6.4→8.9+39%
- 分数Llama 4 MaverickMeta智能指数10.3→9.3-9.7%
- 分数Llama 4 ScoutMeta智能指数4.6→6.5+41%
- 分数Llama 65BMeta智能指数1→5+400%
- 分数Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA智能指数3→7.4+147%
- 分数Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA智能指数6.6→9+36%
- 分数LongCat 2.0LongCat智能指数25.8→19.7-24%
- 分数LongCat 2.0LongCat智能体指数首次评测→15.9
- 分数LongCat Flash LiteLongCat智能指数11.2→11.5+2.7%
- 分数Magistral Medium 1Mistral智能指数6.7→9.1+36%
- 分数Magistral Small 1Mistral智能指数5→8.2+64%
- 分数Magistral Small 1.2Mistral智能指数5.7→8.6+51%
- 分数Mercury 2Inception智能指数14.9→11.5-23%
- 分数Mercury 2Inception智能体指数4.1→4-2.4%
- 分数Mi:dm K 2.5 ProKorea Telecom智能指数10.4→11+5.8%
- 分数MiMo-V2-Flash (Feb 2026)Xiaomi智能指数26.4→22.4-15%
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能指数18.3→16-13%
- 分数MiMo-V2-Flash (Reasoning)Xiaomi智能指数24.5→20.8-15%
- 分数MiMo-V2-OmniXiaomi智能指数28.1→23.9-15%
- 分数MiMo-V2-Omni-0327Xiaomi智能指数29.4→25.1-15%
- 分数MiMo-V2-ProXiaomi智能指数33.1→28.6-14%
- 分数MiMo-V2.5Xiaomi智能指数28.2→22.3-21%
- 分数MiMo-V2.5-ProXiaomi智能指数32.6→26.4-19%
- 分数MiMo-V2.5-Pro (Non-reasoning)Xiaomi智能指数21.3→18.3-14%
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能指数1→5.7+470%
- 分数MiniCPM5-1B (Non-reasoning)OpenBMB智能指数6→8.7+45%
- 分数MiniCPM5-1B (Reasoning)OpenBMB智能指数6.2→8.8+42%
- 分数MiniCPM5-2BOpenBMB智能指数15→14.3-4.7%
- 分数MiniMax M1 40kMiniMax智能指数8.5→10+18%
- 分数MiniMax M1 80kMiniMax智能指数11.6→11.7+0.9%
- 分数MiniMax-M2MiniMax智能指数21.7→18.6-14%
- 分数MiniMax-M2.1MiniMax智能指数24.6→20.9-15%
- 分数MiniMax-M2.5MiniMax智能指数26.8→22.8-15%
- 分数MiniMax-M2.7MiniMax智能指数30.1→23.2-23%
- 分数MiniMax-M2.7MiniMax智能体指数16.9→16.8-0.6%
- 分数MiniMax-M3MiniMax智能指数35.7→29.6-17%
- 分数MiniMax-M3MiniMax智能体指数31→30.8-0.6%
- 分数Ministral 3 14BMistral智能指数5.5→6+9.1%
- 分数Ministral 3 3BMistral智能指数1.7→4.8+182%
- 分数Ministral 3 8BMistral智能指数3.5→5.5+57%
- 分数Mistral 7B InstructMistral智能指数1→5+400%
- 分数Mistral Large (Feb '24)Mistral智能指数1→5.8+480%
- 分数Mistral Large 2 (Jul '24)Mistral智能指数1.7→6.8+300%
- 分数Mistral Large 2 (Nov '24)Mistral智能指数3.5→7.6+117%
- 分数Mistral Large 3Mistral智能指数11.1→9.7-13%
- 分数Mistral Large 3Mistral智能体指数2.5→2.4-4.0%
- 分数Mistral MediumMistral智能指数1→5.5+450%
- 分数Mistral Medium 3Mistral智能指数6.7→9+34%
- 分数Mistral Medium 3.1Mistral智能指数8.4→9.9+18%
- 分数Mistral Medium 3.1Mistral智能体指数3.3→3.1-6.1%
- 分数Mistral Medium 3.5Mistral智能指数22.7→14.9-34%
- 分数Mistral SabaMistral智能指数1→6.5+550%
- 分数Mistral Small (Feb '24)Mistral智能指数1→5.5+450%
- 分数Mistral Small (Sep '24)Mistral智能指数1→5.8+480%
- 分数Mistral Small 3Mistral智能指数1.4→6.7+379%
- 分数Mistral Small 3.1Mistral智能指数10.3→7.4-28%
- 分数Mistral Small 3.2Mistral智能指数5→7+40%
- 分数Mistral Small 4 (Non-reasoning)Mistral智能指数6.6→9+36%
- 分数Mistral Small 4 (Reasoning)Mistral智能指数12.8→11.5-10%
- 分数Mistral Small 4 (Reasoning)Mistral智能体指数1.5→1.4-6.7%
- 分数Mixtral 8x22B InstructMistral智能指数1→5.7+470%
- 分数Mixtral 8x7B InstructMistral智能指数1→5.1+410%
- 分数Molmo 7B-DAllen Institute for AI智能指数1→5.6+460%
- 分数Molmo2-8BAllen Institute for AI智能指数1→5+400%
- 分数Motif 3Motif Technologies智能指数38→33.6-12%
- 分数Motif 3 (Beta)Motif Technologies智能指数36.8→32.3-12%
- 分数Motif-2-12.7B-ReasoningMotif Technologies智能指数7→9.2+31%
- 分数Muse Glimmer (high)Meta智能指数24.4→18.1-26%
- 分数Muse SparkMeta智能指数35.8→31.3-13%
- 分数Muse Spark 1.1 (xhigh)Meta智能指数41.2→34.3-17%
- 分数Muse Spark 1.1 (xhigh)Meta智能体指数27.8→27.5-1.1%
- 分数Muse Spark 1.2 (xhigh)Meta智能指数46.8→39.8-15%
- 分数Muse Spark 1.2 (xhigh)Meta智能体指数44.2→44-0.5%
- 分数Muse Spark 1.3 (xhigh)Meta智能指数51.6→45.2-12%
- 分数Muse Spark 1.3 (xhigh)Meta智能体指数52.1→51.8-0.6%
- 分数Nanbeige4.1-3BNanbeige智能指数5.3→8.4+58%
- 分数Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA智能指数9→10.3+14%
- 分数Nemotron 3 Super 120B A12B (Reasoning)NVIDIA智能指数18.6→13.6-27%
- 分数Nemotron 3 Super 120B A12B (Reasoning)NVIDIA智能体指数4.2→4.1-2.4%
- 分数Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA智能指数29.3→23.4-20%
- 分数Nemotron 3.5 LightningNVIDIA智能指数16.4→13.6-17%
- 分数Nemotron 3.5 LightningNVIDIA智能体指数首次评测→6.1
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能指数11.5→11.7+1.7%
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能指数32.7→28.2-14%
- 分数North Mini CodeCohere智能指数13.5→12.8-5.2%
- 分数Nova 2.0 Lite (high)Amazon智能指数14.3→13.4-6.3%
- 分数Nova 2.0 Lite (low)Amazon智能指数11.7→11.8+0.9%
- 分数Nova 2.0 Lite (medium)Amazon智能指数12.9→12.5-3.1%
- 分数Nova 2.0 Lite (Non-reasoning)Amazon智能指数6→8.7+45%
- 分数Nova 2.0 Omni (low)Amazon智能指数10.6→11.1+4.7%
- 分数Nova 2.0 Omni (medium)Amazon智能指数14.7→13.6-7.5%
- 分数Nova 2.0 Omni (Non-reasoning)Amazon智能指数4.8→8.2+71%
- 分数Nova 2.0 Pro Preview (low)Amazon智能指数13.4→12.8-4.5%
- 分数Nova 2.0 Pro Preview (medium)Amazon智能指数15.5→14.2-8.4%
- 分数Nova 2.0 Pro Preview (Non-reasoning)Amazon智能指数8.5→10+18%
- 分数Nova LiteAmazon智能指数1.4→6.7+379%
- 分数Nova MicroAmazon智能指数1→5.9+490%
- 分数Nova PremierAmazon智能指数6.9→9.2+33%
- 分数Nova ProAmazon智能指数2.1→7+233%
- 分数NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA智能指数1.8→6.8+278%
- 分数NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA智能指数8.6→8.9+3.5%
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能指数3.1→7.4+139%
- 分数NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA智能指数1→5.8+480%
- 分数NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA智能指数3.3→7.5+127%
- 分数NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA智能指数1.8→6.8+278%
- 分数NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA智能指数3.2→7.4+131%
- 分数o1OpenAI智能指数17.1→15.2-11%
- 分数o1-miniOpenAI智能指数8.1→9.8+21%
- 分数o1-previewOpenAI智能指数11→11.4+3.6%
- 分数o1-proOpenAI智能指数12.8→12.4-3.1%
- 分数o3OpenAI智能指数23.7→20.2-15%
- 分数o3-miniOpenAI智能指数12.9→12.5-3.1%
- 分数o3-mini (high)OpenAI智能指数9.6→11+15%
- 分数o3-proOpenAI智能指数25.7→21.9-15%
- 分数o4-mini (high)OpenAI智能指数19.1→16.7-13%
- 分数OLMo 2 32BAllen Institute for AI智能指数1→6+500%
- 分数OLMo 2 7BAllen Institute for AI智能指数1→5.6+460%
- 分数Olmo 3 32B ThinkAllen Institute for AI智能指数1→6.5+550%
- 分数Olmo 3 7B InstructAllen Institute for AI智能指数1→5.2+420%
- 分数Olmo 3 7B ThinkAllen Institute for AI智能指数1→5.6+460%
- 分数Olmo 3.1 32B InstructAllen Institute for AI智能指数1→6.5+550%
- 分数Olmo 3.1 32B ThinkAllen Institute for AI智能指数2.5→7.1+184%
- 分数OpenChat 3.5 (1210)OpenChat智能指数1→5.3+430%
- 分数PALM-2Google智能指数1→5.4+440%
- 分数Phi-3 Mini Instruct 3.8BMicrosoft智能指数1→5.8+480%
- 分数Phi-4Microsoft智能指数1→5.9+490%
- 分数Phi-4 Mini InstructMicrosoft智能指数1→6.3+530%
- 分数Phi-4 Multimodal InstructMicrosoft智能指数1→5.8+480%
- 分数Pixtral LargeMistral智能指数2.5→7.1+184%
- 分数Quasar 438B (max, based on GLM-5.2)Multiverse Computing智能指数34.2→27.1-21%
- 分数Quasar 438B (max, based on GLM-5.2)Multiverse Computing智能体指数32.8→32.7-0.3%
- 分数Qwen Chat 14BAlibaba智能指数1→5+400%
- 分数Qwen Chat 72BAlibaba智能指数1→5.4+440%
- 分数Qwen1.5 Chat 110BAlibaba智能指数1→5.7+470%
- 分数Qwen2 Instruct 72BAlibaba智能指数1→6.3+530%
- 分数Qwen2.5 Coder Instruct 32BAlibaba智能指数1.6→6.7+319%
- 分数Qwen2.5 Coder Instruct 7BAlibaba智能指数1→5.8+480%
- 分数Qwen2.5 Instruct 32BAlibaba智能指数1.9→6.9+263%
- 分数Qwen2.5 Instruct 72BAlibaba智能指数3.9→7.7+97%
- 分数Qwen2.5 MaxAlibaba智能指数4.5→8+78%
- 分数Qwen2.5 TurboAlibaba智能指数1→6.4+540%
- 分数Qwen3 0.6B (Non-reasoning)Alibaba智能指数1→4.8+380%
- 分数Qwen3 0.6B (Reasoning)Alibaba智能指数1→4.8+380%
- 分数Qwen3 1.7B (Non-reasoning)Alibaba智能指数1→4.9+390%
- 分数Qwen3 1.7B (Reasoning)Alibaba智能指数1→5.2+420%
- 分数Qwen3 14B (Non-reasoning)Alibaba智能指数1.5→6.7+347%
- 分数Qwen3 14B (Reasoning)Alibaba智能指数7.1→6.4-9.9%
- 分数Qwen3 235B A22B (Non-reasoning)Alibaba智能指数5.1→8.3+63%
- 分数Qwen3 235B A22B (Reasoning)Alibaba智能指数7.6→9.5+25%
- 分数Qwen3 235B A22B 2507 (Reasoning)Alibaba智能指数13.1→12.7-3.1%
- 分数Qwen3 235B A22B 2507 InstructAlibaba智能指数12.1→12-0.8%
- 分数Qwen3 30B A3B (Non-reasoning)Alibaba智能指数1.3→6.6+408%
- 分数Qwen3 30B A3B (Reasoning)Alibaba智能指数3.6→7.6+111%
- 分数Qwen3 30B A3B 2507 (Reasoning)Alibaba智能指数8.6→9.8+14%
- 分数Qwen3 30B A3B 2507 InstructAlibaba智能指数3.4→7.5+121%
- 分数Qwen3 32B (Non-reasoning)Alibaba智能指数3→7.3+143%
- 分数Qwen3 32B (Reasoning)Alibaba智能指数8→7.2-10.0%
- 分数Qwen3 4B (Non-reasoning)Alibaba智能指数1.2→6.6+450%
- 分数Qwen3 4B (Reasoning)Alibaba智能指数2.7→7.2+167%
- 分数Qwen3 4B 2507 (Reasoning)Alibaba智能指数6.2→8.8+42%
- 分数Qwen3 4B 2507 InstructAlibaba智能指数1.6→6.7+319%
- 分数Qwen3 8B (Non-reasoning)Alibaba智能指数1→6+500%
- 分数Qwen3 8B (Reasoning)Alibaba智能指数2.8→5.2+86%
- 分数Qwen3 Coder 30B A3B InstructAlibaba智能指数7.7→9.6+25%
- 分数Qwen3 Coder NextAlibaba智能指数14.1→10.1-28%
- 分数Qwen3 Coder NextAlibaba智能体指数3.7→3.6-2.7%
- 分数Qwen3 MaxAlibaba智能指数17.6→15.6-11%
- 分数Qwen3 Max (Preview)Alibaba智能指数13→12.6-3.1%
- 分数Qwen3 Max ThinkingAlibaba智能指数25→21.3-15%
- 分数Qwen3 Max Thinking (Preview)Alibaba智能指数18.6→16.3-12%
- 分数Qwen3 Next 80B A3B (Reasoning)Alibaba智能指数10.7→11.2+4.7%
- 分数Qwen3 Next 80B A3B InstructAlibaba智能指数7.8→9.6+23%
- 分数Qwen3 Omni 30B A3B (Reasoning)Alibaba智能指数4→7.8+95%
- 分数Qwen3 Omni 30B A3B InstructAlibaba智能指数1→6+500%
- 分数Qwen3 VL 235B A22B (Reasoning)Alibaba智能指数14.4→13.4-6.9%
- 分数Qwen3 VL 235B A22B InstructAlibaba智能指数8.4→9.9+18%
- 分数Qwen3 VL 30B A3B (Reasoning)Alibaba智能指数7.5→9.5+27%
- 分数Qwen3 VL 30B A3B InstructAlibaba智能指数4.3→7.9+84%
- 分数Qwen3 VL 32B InstructAlibaba智能指数5.3→8.4+58%
- 分数Qwen3 VL 4B (Reasoning)Alibaba智能指数2.3→7+204%
- 分数Qwen3 VL 4B InstructAlibaba智能指数1→5.7+470%
- 分数Qwen3 VL 8B (Reasoning)Alibaba智能指数4.9→8.2+67%
- 分数Qwen3 VL 8B InstructAlibaba智能指数2.8→7.3+161%
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能指数1→5.4+440%
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能指数1→6.1+510%
- 分数Qwen3.5 122B A10B (Non-reasoning)Alibaba智能指数20.6→17.7-14%
- 分数Qwen3.5 122B A10B (Reasoning)Alibaba智能指数24.8→16.2-35%
- 分数Qwen3.5 122B A10B (Reasoning)Alibaba智能体指数9.8→9.6-2.0%
- 分数Qwen3.5 27B (Non-reasoning)Alibaba智能指数22.7→19.4-15%
- 分数Qwen3.5 27B (Reasoning)Alibaba智能指数26.9→22.9-15%
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能指数1→6.2+520%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数2.1→6.9+229%
- 分数Qwen3.5 35B A3B (Non-reasoning)Alibaba智能指数17→15.1-11%
- 分数Qwen3.5 35B A3B (Reasoning)Alibaba智能指数22.6→19.3-15%
- 分数Qwen3.5 397B A17B (Non-reasoning)Alibaba智能指数25.2→21.4-15%
- 分数Qwen3.5 397B A17B (Reasoning)Alibaba智能指数26.1→19.1-27%
- 分数Qwen3.5 4B (Non-reasoning)Alibaba智能指数10→10.8+8.0%
- 分数Qwen3.5 4B (Reasoning)Alibaba智能指数13.9→13.1-5.8%
- 分数Qwen3.5 9B (Non-reasoning)Alibaba智能指数14.1→13.3-5.7%
- 分数Qwen3.5 9B (Reasoning)Alibaba智能指数14.8→13.7-7.4%
- 分数Qwen3.5 Omni FlashAlibaba智能指数12.9→12.5-3.1%
- 分数Qwen3.5 Omni PlusAlibaba智能指数24→20.4-15%
- 分数Qwen3.6 27B (Non-reasoning)Alibaba智能指数23.3→19.8-15%
- 分数Qwen3.6 27B (Reasoning)Alibaba智能指数29→21.9-24%
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能指数17.1→15.2-11%
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能指数26.2→22.3-15%
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能体指数15.1→15-0.7%
- 分数Qwen3.6 Max PreviewAlibaba智能指数32.9→28.4-14%
- 分数Qwen3.6 PlusAlibaba智能指数31.5→27-14%
- 分数Qwen3.7 MaxAlibaba智能指数36.6→29.9-18%
- 分数Qwen3.7 PlusAlibaba智能指数31.9→25.8-19%
- 分数Qwen3.7 PlusAlibaba智能体指数19.8→19.7-0.5%
- 分数Qwen3.8 2.4T A95BAlibaba智能指数46.7→40-14%
- 分数Qwen3.8 2.4T A95BAlibaba智能体指数50.7→50.4-0.6%
- 分数Qwen3.8 27B (low)Alibaba智能指数33.8→29.2-14%
- 分数Qwen3.8 27B (low)Alibaba智能体指数首次评测→39.9
- 分数Qwen3.8 27B (medium)Alibaba智能指数35.2→30.7-13%
- 分数Qwen3.8 27B (Non-reasoning)Alibaba智能指数26.5→22.4-15%
- 分数Qwen3.8 27B (xhigh)Alibaba智能指数41.4→33.9-18%
- 分数Qwen3.8 27B (xhigh)Alibaba智能体指数46.8→46.5-0.6%
- 分数Qwen3.8 MaxAlibaba智能指数46.9→40.3-14%
- 分数Qwen3.8 MaxAlibaba智能体指数49.9→49.6-0.6%
- 分数Qwen3.8-Flash-NextAlibaba智能指数45.6→42.2-7.5%
- 分数QwQ 32BAlibaba智能指数7.5→9.5+27%
- 分数QwQ 32B-PreviewAlibaba智能指数3.6→7.6+111%
- 分数R1 1776Perplexity智能指数1→6.4+540%
- 分数Reka Flash (Sep '24)Reka AI智能指数1→6.4+540%
- 分数Reka Flash 3Reka AI智能指数1→5.6+460%
- 分数Ring-1TInclusionAI智能指数10.2→10.9+6.9%
- 分数Ring-2.6-1TInclusionAI智能指数23.7→17.3-27%
- 分数Ring-2.6-1TInclusionAI智能体指数13→12.9-0.8%
- 分数Ring-flash-2.0InclusionAI智能指数2.6→7.2+177%
- 分数Sarvam 105B (high)Sarvam智能指数6.2→8.8+42%
- 分数Sarvam 30B (high)Sarvam智能指数1.1→6.6+500%
- 分数Sarvam M (Reasoning, based on Mistral Small 3.1)Sarvam智能指数1→5.3+430%
- 分数Seed-OSS-36B-InstructByteDance Seed智能指数12.2→12.1-0.8%
- 分数Solar MiniUpstage智能指数1→6.4+540%
- 分数Solar Open 100B (Reasoning)Upstage智能指数9.2→10.4+13%
- 分数Solar Open2 250BUpstage智能指数29→24.7-15%
- 分数Solar Pro 2 (Non-reasoning)Upstage智能指数2.2→7+218%
- 分数Solar Pro 2 (Preview) (Non-reasoning)Upstage智能指数4.3→7.9+84%
- 分数Solar Pro 2 (Preview) (Reasoning)Upstage智能指数6.7→9.1+36%
- 分数Solar Pro 2 (Reasoning)Upstage智能指数3.3→7.5+127%
- 分数Solar Pro 3Upstage智能指数9.4→7.8-17%
- 分数Solar Pro 4Upstage智能指数32.7→28.2-14%
- 分数SonarPerplexity智能指数3.8→7.7+103%
- 分数Sonar ProPerplexity智能指数3.6→7.6+111%
- 分数Sonar ReasoningPerplexity智能指数5.9→8.7+47%
- 分数Step 3.5 FlashStepFun智能指数19.1→16.6-13%
- 分数Step 3.5 Flash 2603StepFun智能指数19.5→17-13%
- 分数Step 3.7 FlashStepFun智能指数22.9→19.5-15%
- 分数Step3 VL 10BStepFun智能指数3.8→7.7+103%
- 分数Tiny Aya GlobalCohere智能指数1→4.8+380%
- 分数Tri-21B-ThinkTrillion Labs智能指数6.6→9+36%
- 分数Tri-21B-think PreviewTrillion Labs智能指数7.7→9.6+25%
- 分数Trinity Large ThinkingArcee AI智能指数13.1→10.9-17%
09-08
2026年9月8日1 条变化- 新增MiniCPM5-2BOpenBMB智能指数 15
09-06
2026年9月6日6 条变化- 分数Gemini 3.1 Pro PreviewGoogle智能指数38.3→36.7-4.2%
- 分数Gemini 3.1 Pro PreviewGoogle编程指数68.7→68.8+0.1%
- 分数Mistral Large 3Mistral智能指数9.7→11.1+14%
- 分数Mistral Large 3Mistral编程指数20.2→20.1-0.5%
- 分数Muse Spark 1.3 (max)Meta智能指数52.7→53+0.6%
- 分数Qwen3.8 27B (xhigh)Alibaba智能指数41.6→41.4-0.5%
09-05
2026年9月5日20 条变化- 价格Command-R (Mar '24)Cohere输入价$0.500→指标撤下
- 价格Command-R (Mar '24)Cohere输出价$1.50→指标撤下
- 价格Command-R+ (Apr '24)Cohere输入价$3.00→指标撤下
- 价格Command-R+ (Apr '24)Cohere输出价$15.00→指标撤下
- 价格Gemma 4 31B (Non-reasoning)Google输入价$0.150→$0.140-6.7%
- 价格GLM-5.1 (Reasoning)Z AI输入价$1.28→$1.20-6.3%
- 价格GLM-5.1 (Reasoning)Z AI输出价$4.07→$4.40+8.1%
- 价格gpt-oss-120b (high)OpenAI输出价$0.600→$0.590-1.7%
- 价格gpt-oss-120b (low)OpenAI输出价$0.590→$0.540-8.5%
- 价格Kimi K2.5 (Reasoning)Kimi输出价$3.00→$2.75-8.3%
- 价格Llama 4 ScoutMeta输入价$0.180→$0.190+5.6%
- 价格Llama 4 ScoutMeta输出价$0.660→$0.680+3.0%
- 价格Muse Glimmer (high)Meta输入价$0.320→$0.350+9.4%
- 价格Muse Glimmer (high)Meta输出价$1.35→$1.50+11%
- 价格Muse Spark 1.3 (max)Meta输入价首次评测→$1.25
- 价格Muse Spark 1.3 (max)Meta输出价首次评测→$4.25
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输入价$0.680→$0.600-12%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输出价$2.67→$2.60-2.6%
- 价格Nemotron 3.5 LightningNVIDIA输入价$0.070→$0.060-14%
- 价格Nemotron 3.5 LightningNVIDIA输出价$0.220→$0.200-9.1%
09-04
2026年9月4日9 条变化- 新增GPT-6 Astra (high)OpenAI智能指数 60.3
- 新增GPT-6 Astra (low)OpenAI智能指数 56.7
- 新增GPT-6 Astra (max)OpenAI智能指数 61.2
- 新增GPT-6 Astra (medium)OpenAI智能指数 59.2
- 新增GPT-6 Astra (Non-reasoning)OpenAI智能指数 55.3
- 新增GPT-6 Astra (xhigh)OpenAI智能指数 61
- 新增K2 Horizon 375B A23BMBZUAI Institute of Foundation Models智能指数 47.3
- 新增Muse Spark 1.3 (max)Meta智能指数 62.1
- 新增Muse Spark 1.3 (xhigh)Meta智能指数 60.8
09-03
2026年9月3日3 条变化- 新增Gemini 3.8 Flash (high)Google智能指数 58.7
- 新增Gemini 3.8 Flash (low)Google智能指数 51.7
- 新增Gemini 3.8 Flash (medium)Google智能指数 56.6
09-02
2026年9月2日11 条变化- 新增Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic智能指数 62.5
- 新增Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic智能指数 58.1
- 新增Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic智能指数 65.7
- 新增Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic智能指数 60.5
- 新增Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic智能指数 64.8
- 新增NVIDIA Nemotron 3 Nano 4BNVIDIA智能指数 8.6
- 新增Quasar 438B (max, based on GLM-5.2)Multiverse Computing智能指数 43
- 价格Gemma 3n E4B InstructGoogle输入价$0.060→指标撤下
- 价格Gemma 3n E4B InstructGoogle输出价$0.120→指标撤下
- 价格Trinity Large ThinkingArcee AI输入价$0.230→$0.250+8.7%
- 价格Trinity Large ThinkingArcee AI输出价$0.880→$0.900+2.3%
09-01
2026年9月1日7 条变化- 新增Apodex 1.1Apodex智能指数 44
- 下架NVIDIA Nemotron 3 Nano 4BNVIDIA
- 价格GLM-5.1 (Non-reasoning)Z AI输入价$1.39→$1.38-0.7%
- 价格GLM-5.1 (Reasoning)Z AI输入价$1.38→$1.28-7.2%
- 价格GLM-5.1 (Reasoning)Z AI输出价$4.40→$4.07-7.5%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输入价$0.600→$0.680+13%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输出价$2.75→$2.67-2.9%
08-28
2026年8月28日3 条变化- 新增Agnes 2.5 Pro BetaSapiens AI智能指数 49.1
- 新增Granite 4.2 30BIBM智能指数 23.7
- 新增Qwen3.8-Flash-NextAlibaba智能指数 55.8
08-27
2026年8月27日9 条变化- 新增DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek智能指数 51.5
- 新增GLM 5.3 FlashZ AI智能指数 57.5
- 新增Granite 4.2 3BIBM智能指数 14.3
- 新增Granite 4.2 8BIBM智能指数 19.6
- 新增Qwen3.8 27B (Non-reasoning)Alibaba智能指数 34.7
- 价格DeepSeek V3.1 (Non-reasoning)DeepSeek输入价$0.560→$0.570+1.8%
- 价格GLM-5.1 (Reasoning)Z AI输入价$1.39→$1.38-0.7%
- 价格HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing输入价首次评测→$0.040
- 价格HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing输出价首次评测→$0.140
08-25
2026年8月25日25 条变化- 新增G9v3-39A5BAI9Stars智能指数 34
- 新增Granite 4.0 1BIBM智能指数 1.6
- 新增Grok 4.6 (low)SpaceXAI智能指数 51.7
- 新增Grok 4.6 (medium)SpaceXAI智能指数 59
- 新增Grok 4.6 (xhigh)SpaceXAI智能指数 60
- 新增K2-V2 (high)MBZUAI Institute of Foundation Models智能指数 14.2
- 新增LFM2.5-2.6BLiquid AI智能指数 11
- 新增Motif 3 (Beta)Motif Technologies智能指数 45.3
- 下架G9v3-39A5BAI9Stars
- 价格GPT-5.6 Sol (high)OpenAI输入价$5.00→$4.00-20%
- 价格GPT-5.6 Sol (high)OpenAI输出价$30.00→$20.00-33%
- 价格GPT-5.6 Sol (low)OpenAI输入价$5.00→$4.00-20%
- 价格GPT-5.6 Sol (low)OpenAI输出价$30.00→$20.00-33%
- 价格GPT-5.6 Sol (max)OpenAI输入价$5.00→$4.00-20%
- 价格GPT-5.6 Sol (max)OpenAI输出价$30.00→$20.00-33%
- 价格GPT-5.6 Sol (medium)OpenAI输入价$5.00→$4.00-20%
- 价格GPT-5.6 Sol (medium)OpenAI输出价$30.00→$20.00-33%
- 价格GPT-5.6 Sol (Non-reasoning)OpenAI输入价$5.00→$4.00-20%
- 价格GPT-5.6 Sol (Non-reasoning)OpenAI输出价$30.00→$20.00-33%
- 价格GPT-5.6 Sol (xhigh)OpenAI输入价$5.00→$4.00-20%
- 价格GPT-5.6 Sol (xhigh)OpenAI输出价$30.00→$20.00-33%
- 价格Ling 2.6 FlashInclusionAI输入价$0.100→指标撤下
- 价格Ling 2.6 FlashInclusionAI输出价$0.300→指标撤下
- 价格Qwen3.8 27B (xhigh)Alibaba输入价$0.450→$0.500+11%
- 价格Qwen3.8 27B (xhigh)Alibaba输出价$3.20→$3.00-6.3%
08-24
2026年8月24日9 条变化- 新增Qwen3.8 27B (low)Alibaba智能指数 42.9
- 新增Qwen3.8 27B (medium)Alibaba智能指数 44.5
- 下架Granite 4.0 1BIBM
- 下架K2-V2 (high)MBZUAI Institute of Foundation Models
- 下架Motif 3 (Beta)Motif Technologies
- 价格Llama 4 MaverickMeta输入价$0.270→$0.260-3.7%
- 价格Llama 4 MaverickMeta输出价$0.850→$0.910+7.1%
- 价格Magistral Medium 1.2Mistral输入价$2.00→指标撤下
- 价格Magistral Medium 1.2Mistral输出价$5.00→指标撤下
08-20
2026年8月20日39 条变化- 新增GLM-5.3 (max)Z AI智能指数 59.5
- 新增Qwen3.8 27B (xhigh)Alibaba智能指数 52
- 价格DeepSeek R1 (Jan '25)DeepSeek输入价$1.68→$2.00+19%
- 价格DeepSeek R1 (Jan '25)DeepSeek输出价$4.70→$4.00-15%
- 价格DeepSeek R1 0528 (May '25)DeepSeek输出价$4.20→$3.00-29%
- 价格DeepSeek V4 Flash 0420 (Non-reasoning)DeepSeek输入价$0.140→$0.090-36%
- 价格DeepSeek V4 Flash 0420 (Non-reasoning)DeepSeek输出价$0.280→$0.200-29%
- 价格DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek输入价$0.140→$0.130-7.1%
- 价格Gemini 3.6 Flash (high)Google输入价$1.50→$0.750-50%
- 价格Gemini 3.6 Flash (high)Google输出价$7.50→$3.75-50%
- 价格Gemma 4 26B A4B (Reasoning)Google输入价$0.130→$0.120-7.7%
- 价格Gemma 4 26B A4B (Reasoning)Google输出价$0.400→$0.370-7.5%
- 价格GLM-5.1 (Non-reasoning)Z AI输入价$1.38→$1.39+0.7%
- 价格GLM-5.1 (Reasoning)Z AI输入价$1.38→$1.39+0.7%
- 价格Hy3Tencent输出价$0.560→$0.550-1.8%
- 价格Nemotron 3.5 LightningNVIDIA输入价$0.050→$0.070+40%
- 价格Nemotron 3.5 LightningNVIDIA输出价$0.200→$0.220+10.0%
- 价格Qwen3 235B A22B 2507 (Reasoning)Alibaba输入价$0.700→$0.230-67%
- 价格Qwen3 235B A22B 2507 (Reasoning)Alibaba输出价$8.40→$2.30-73%
- 价格Qwen3 235B A22B 2507 InstructAlibaba输入价$0.700→$0.230-67%
- 价格Qwen3 235B A22B 2507 InstructAlibaba输出价$2.80→$0.920-67%
- 价格Qwen3 32B (Non-reasoning)Alibaba输入价$0.700→$0.160-77%
- 价格Qwen3 32B (Non-reasoning)Alibaba输出价$2.80→$0.640-77%
- 价格Qwen3 32B (Reasoning)Alibaba输入价$0.700→$0.160-77%
- 价格Qwen3 32B (Reasoning)Alibaba输出价$8.40→$0.640-92%
- 价格Qwen3 Next 80B A3B (Reasoning)Alibaba输入价$0.500→$0.150-70%
- 价格Qwen3 Next 80B A3B (Reasoning)Alibaba输出价$6.00→$1.20-80%
- 价格Qwen3 Next 80B A3B InstructAlibaba输入价$0.500→$0.150-70%
- 价格Qwen3 Next 80B A3B InstructAlibaba输出价$2.00→$1.20-40%
- 价格Qwen3 VL 235B A22B (Reasoning)Alibaba输入价$0.700→$0.400-43%
- 价格Qwen3 VL 235B A22B (Reasoning)Alibaba输出价$8.40→$4.00-52%
- 价格Qwen3 VL 235B A22B InstructAlibaba输入价$0.700→$0.400-43%
- 价格Qwen3 VL 235B A22B InstructAlibaba输出价$2.80→$1.60-43%
- 价格Qwen3 VL 32B (Reasoning)Alibaba输入价$0.700→$0.160-77%
- 价格Qwen3 VL 32B (Reasoning)Alibaba输出价$8.40→$0.640-92%
- 价格Qwen3 VL 32B InstructAlibaba输入价$0.700→$0.160-77%
- 价格Qwen3 VL 32B InstructAlibaba输出价$2.80→$0.640-77%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能指数31.1→31.4+1.0%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能体指数首次评测→19.9
08-16
2026年8月16日3 条变化- 分数LongCat 2.0LongCat智能指数34.3→34-0.9%
- 分数LongCat 2.0LongCat智能体指数首次评测→22
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能体指数首次评测→19.2
08-15
2026年8月15日10 条变化- 价格DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek输入价$0.140→$0.440+214%
- 价格DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek输出价$0.280→$1.32+371%
- 分数KAT Coder Pro V2KwaiKAT智能指数34.5→33.7-2.3%
- 分数KAT Coder Pro V2KwaiKAT智能体指数首次评测→15.1
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能指数42.1→41.7-1.0%
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能体指数首次评测→31.2
- 分数Ring-2.6-1TInclusionAI智能指数31.3→31.7+1.3%
- 分数Ring-2.6-1TInclusionAI智能体指数首次评测→20.1
- 分数Trinity Large ThinkingArcee AI智能指数18.4→18.7+1.6%
- 分数Trinity Large ThinkingArcee AI智能体指数首次评测→3.7
08-14
2026年8月14日8 条变化- 新增Gemini 3.7 Flash (high)Google智能指数 56
- 新增Gemini 3.7 Flash (low)Google智能指数 50.9
- 新增Gemini 3.7 Flash (medium)Google智能指数 53.4
- 新增Qwen3.8 2.4T A95BAlibaba智能指数 57.7
- 下架Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 5 Fallback)Anthropic
- 价格DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek输入价$0.430→$1.32+207%
- 价格DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek输出价$0.870→$3.96+355%
- 分数DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek智能指数53→53.2+0.4%
08-13
2026年8月13日19 条变化- 新增A.X-K2SK Telecom智能指数 35
- 新增Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 5 Fallback)Anthropic
- 新增DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek智能指数 53
- 新增Grok 4.6 (high)SpaceXAI智能指数 60.9
- 新增K-EXAONE 2.0 0803LG AI Research智能指数 31
- 新增Motif 3Motif Technologies智能指数 47.4
- 新增Solar Open2 250BUpstage智能指数 37.4
- 新增Solar Pro 4Upstage智能指数 41.6
- 价格Claude 3 SonnetAnthropic输入价$3.00→指标撤下
- 价格Claude 3 SonnetAnthropic输出价$15.00→指标撤下
- 价格gpt-oss-120b (high)OpenAI输出价$0.590→$0.600+1.7%
- 价格Jamba 1.6 LargeAI21 Labs输入价$2.00→指标撤下
- 价格Jamba 1.6 LargeAI21 Labs输出价$8.00→指标撤下
- 价格Jamba 1.6 MiniAI21 Labs输入价$0.200→指标撤下
- 价格Jamba 1.6 MiniAI21 Labs输出价$0.400→指标撤下
- 价格Jamba 1.7 LargeAI21 Labs输入价$2.00→指标撤下
- 价格Jamba 1.7 LargeAI21 Labs输出价$8.00→指标撤下
- 价格Muse Glimmer (high)Meta输入价首次评测→$0.320
- 价格Muse Glimmer (high)Meta输出价首次评测→$1.35
08-12
2026年8月12日95 条变化- 新增Muse Glimmer (high)Meta智能指数 35.1
- 新增Nemotron 3.5 LightningNVIDIA智能指数 23.6
- 价格DeepSeek R1 Distill Llama 70BDeepSeek输出价$1.05→$1.10+4.8%
- 价格GLM-5.2 (max)Z AI输入价$1.35→$1.40+3.7%
- 价格GLM-5.2 (max)Z AI输出价$4.29→$4.40+2.6%
- 价格GLM-5.2 (Non-reasoning)Z AI输入价$1.35→$1.40+3.7%
- 价格GLM-5.2 (Non-reasoning)Z AI输出价$4.25→$4.40+3.5%
- 价格gpt-oss-120b (low)OpenAI输出价$0.540→$0.590+9.3%
- 价格HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing输入价$0.040→指标撤下
- 价格HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing输出价$0.140→指标撤下
- 价格Llama 3.1 Instruct 8BMeta输入价$0.050→$0.020-60%
- 价格Llama 3.1 Instruct 8BMeta输出价$0.080→$0.050-38%
- 价格Qwen3.5 9B (Non-reasoning)Alibaba输入价首次评测→$0.170
- 价格Qwen3.5 9B (Non-reasoning)Alibaba输出价首次评测→$0.250
- 价格Qwen3.6 35B A3B (Reasoning)Alibaba输入价$0.250→$0.380+52%
- 价格Qwen3.6 35B A3B (Reasoning)Alibaba输出价$1.49→$2.25+51%
- 分数Celeris-1Celeris智能指数12.5→12.4-0.8%
- 分数Celeris-1Celeris智能体指数2.8→2.4-14%
- 分数Claude 3.5 HaikuAnthropic智能指数12.4→12.2-1.6%
- 分数Claude 3.5 HaikuAnthropic智能体指数1.9→指标撤下
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能指数30.6→31.1+1.6%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能体指数18.1→指标撤下
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能指数32.6→32.8+0.6%
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能体指数18.3→指标撤下
- 分数DiffusionGemma 26B A4BGoogle智能指数13.8→13.5-2.2%
- 分数DiffusionGemma 26B A4BGoogle智能体指数2.2→指标撤下
- 分数EXAONE 4.5 33BLG AI Research智能体指数9.7→指标撤下
- 分数G9v3-39A5BAI9Stars智能指数31.4→31.6+0.6%
- 分数G9v3-39A5BAI9Stars智能体指数28.7→指标撤下
- 分数G9v3-3BAI9Stars智能指数16.5→16.2-1.8%
- 分数G9v3-3BAI9Stars智能体指数14→指标撤下
- 分数Gemma 4 12B (Reasoning)Google智能指数22.3→22.2-0.4%
- 分数Gemma 4 12B (Reasoning)Google智能体指数7.9→指标撤下
- 分数Gemma 4 E2B (Reasoning)Google智能指数9.8→9.5-3.1%
- 分数Gemma 4 E2B (Reasoning)Google智能体指数1.5→指标撤下
- 分数Gemma 4 E4B (Reasoning)Google智能指数12.4→12.2-1.6%
- 分数Gemma 4 E4B (Reasoning)Google智能体指数1.8→指标撤下
- 分数Granite 4.1 30BIBM智能指数8.9→8.7-2.2%
- 分数Granite 4.1 30BIBM智能体指数1.3→指标撤下
- 分数Granite 4.1 3BIBM智能指数4.7→4.4-6.4%
- 分数Granite 4.1 3BIBM智能体指数0.4→指标撤下
- 分数JT-4.1 Flash 236B A21BChina Mobile智能指数39.8→39.9+0.3%
- 分数JT-4.1 Flash 236B A21BChina Mobile智能体指数34.9→指标撤下
- 分数K-EXAONE (Reasoning)LG AI Research智能指数22.4→22.5+0.4%
- 分数K-EXAONE (Reasoning)LG AI Research智能体指数8→指标撤下
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能指数17.2→17.4+1.2%
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能体指数1.8→指标撤下
- 分数KAT Coder Pro V2KwaiKAT智能指数33.9→34.5+1.8%
- 分数KAT Coder Pro V2KwaiKAT智能体指数15.5→指标撤下
- 分数Ling 3.0 TinyInclusionAI智能指数24.3→24.5+0.8%
- 分数Ling 3.0 TinyInclusionAI智能体指数15.4→16+3.9%
- 分数Llama 3.1 Instruct 8BMeta智能指数7.8→7.4-5.1%
- 分数Llama 3.1 Instruct 8BMeta智能体指数0.5→指标撤下
- 分数Llama 3.3 Instruct 70BMeta智能体指数0.3→指标撤下
- 分数LongCat 2.0LongCat智能指数33.9→34.3+1.2%
- 分数LongCat 2.0LongCat智能体指数21.8→指标撤下
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能指数25→25.1+0.4%
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能体指数12→指标撤下
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能指数4.3→3.8-12%
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能体指数1.4→指标撤下
- 分数Motif 3 (Beta)Motif Technologies智能指数44.9→45.3+0.9%
- 分数Motif 3 (Beta)Motif Technologies智能体指数34.9→指标撤下
- 分数Muse SparkMeta智能指数43.5→44.3+1.8%
- 分数Muse SparkMeta智能体指数28.7→指标撤下
- 分数Nanbeige4.1-3BNanbeige智能指数11.2→11-1.8%
- 分数Nanbeige4.1-3BNanbeige智能体指数0→指标撤下
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能指数18→17.8-1.1%
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能体指数3.7→指标撤下
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能指数41.7→42.1+1.0%
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能体指数31→指标撤下
- 分数Nova 2.0 Lite (high)Amazon智能指数18.4→20.8+13%
- 分数Nova 2.0 Lite (high)Amazon智能体指数3.1→指标撤下
- 分数Nova 2.0 Pro Preview (low)Amazon智能指数20→19.8-1.0%
- 分数Nova 2.0 Pro Preview (low)Amazon智能体指数6.4→指标撤下
- 分数Nova 2.0 Pro Preview (medium)Amazon智能指数22→22.1+0.5%
- 分数Nova 2.0 Pro Preview (medium)Amazon智能体指数7→指标撤下
- 分数Nova 2.0 Pro Preview (Non-reasoning)Amazon智能体指数2.9→指标撤下
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能指数8.8→8.6-2.3%
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能体指数2.3→指标撤下
- 分数Phi-4 Mini InstructMicrosoft智能指数6.2→5.7-8.1%
- 分数Phi-4 Mini InstructMicrosoft智能体指数0.3→指标撤下
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能指数3.5→2.9-17%
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能体指数0.4→指标撤下
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能指数5→5.2+4.0%
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能体指数5.7→指标撤下
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能指数5.7→5.3-7.0%
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能体指数0.8→指标撤下
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数7.5→7.4-1.3%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能体指数0.8→指标撤下
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能指数24.5→24.6+0.4%
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能体指数19.1→指标撤下
- 分数Ring-2.6-1TInclusionAI智能指数31.1→31.3+0.6%
- 分数Ring-2.6-1TInclusionAI智能体指数18.9→指标撤下
- 分数Trinity Large ThinkingArcee AI智能指数18.6→18.4-1.1%
- 分数Trinity Large ThinkingArcee AI智能体指数3.7→指标撤下
08-08
2026年8月8日85 条变化- 价格GLM-5.2 (max)Z AI输入价$1.30→$1.35+3.8%
- 价格GLM-5.2 (max)Z AI输出价$4.18→$4.29+2.6%
- 价格gpt-oss-120b (low)OpenAI输出价$0.590→$0.540-8.5%
- 分数Celeris-1Celeris智能指数12.4→12.5+0.8%
- 分数Celeris-1Celeris智能体指数2.4→2.8+17%
- 分数Claude 3.5 HaikuAnthropic智能指数12.2→12.4+1.6%
- 分数Claude 3.5 HaikuAnthropic智能体指数首次评测→1.9
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能指数31.1→30.6-1.6%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能体指数首次评测→18.1
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能指数32.8→32.6-0.6%
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能体指数首次评测→18.3
- 分数DiffusionGemma 26B A4BGoogle智能指数13.5→13.8+2.2%
- 分数DiffusionGemma 26B A4BGoogle智能体指数首次评测→2.2
- 分数EXAONE 4.5 33BLG AI Research智能体指数首次评测→9.7
- 分数G9v3-39A5BAI9Stars智能指数31.6→31.4-0.6%
- 分数G9v3-39A5BAI9Stars智能体指数首次评测→28.7
- 分数G9v3-3BAI9Stars智能指数16.2→16.5+1.9%
- 分数G9v3-3BAI9Stars智能体指数首次评测→14
- 分数Gemma 4 12B (Reasoning)Google智能指数22.2→22.3+0.5%
- 分数Gemma 4 12B (Reasoning)Google智能体指数首次评测→7.9
- 分数Gemma 4 31B (Reasoning)Google智能指数30→29.7-1.0%
- 分数Gemma 4 31B (Reasoning)Google智能体指数首次评测→14.4
- 分数Gemma 4 E2B (Reasoning)Google智能指数9.5→9.8+3.2%
- 分数Gemma 4 E2B (Reasoning)Google智能体指数首次评测→1.5
- 分数Gemma 4 E4B (Reasoning)Google智能指数12.2→12.4+1.6%
- 分数Gemma 4 E4B (Reasoning)Google智能体指数首次评测→1.8
- 分数Granite 4.1 30BIBM智能指数8.7→8.9+2.3%
- 分数Granite 4.1 30BIBM智能体指数首次评测→1.3
- 分数Granite 4.1 3BIBM智能指数4.4→4.7+6.8%
- 分数Granite 4.1 3BIBM智能体指数首次评测→0.4
- 分数JT-4.1 Flash 236B A21BChina Mobile智能指数39.9→39.8-0.3%
- 分数JT-4.1 Flash 236B A21BChina Mobile智能体指数首次评测→34.9
- 分数K-EXAONE (Reasoning)LG AI Research智能指数22.5→22.4-0.4%
- 分数K-EXAONE (Reasoning)LG AI Research智能体指数首次评测→8
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能指数17.4→17.2-1.1%
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能体指数首次评测→1.8
- 分数KAT Coder Pro V2KwaiKAT智能指数34.5→33.9-1.7%
- 分数KAT Coder Pro V2KwaiKAT智能体指数首次评测→15.5
- 分数Ling 3.0 TinyInclusionAI智能指数23.9→24.3+1.7%
- 分数Ling 3.0 TinyInclusionAI编程指数25.8→26.5+2.7%
- 分数Ling 3.0 TinyInclusionAI智能体指数首次评测→15.4
- 分数Llama 3.1 Instruct 8BMeta智能指数7.4→7.8+5.4%
- 分数Llama 3.1 Instruct 8BMeta智能体指数首次评测→0.5
- 分数Llama 3.3 Instruct 70BMeta智能体指数首次评测→0.3
- 分数LongCat 2.0LongCat智能指数34.3→33.9-1.2%
- 分数LongCat 2.0LongCat智能体指数首次评测→21.8
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能指数25.1→25-0.4%
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能体指数首次评测→12
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能指数3.8→4.3+13%
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能体指数首次评测→1.4
- 分数Motif 3 (Beta)Motif Technologies智能指数45.3→44.9-0.9%
- 分数Motif 3 (Beta)Motif Technologies智能体指数首次评测→34.9
- 分数Muse SparkMeta智能指数44.3→43.5-1.8%
- 分数Muse SparkMeta智能体指数首次评测→28.7
- 分数Nanbeige4.1-3BNanbeige智能指数11→11.2+1.8%
- 分数Nanbeige4.1-3BNanbeige智能体指数首次评测→0
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能指数17.8→18+1.1%
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能体指数首次评测→3.7
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能指数42.1→41.7-1.0%
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能体指数首次评测→31
- 分数Nova 2.0 Lite (high)Amazon智能指数20.8→18.4-12%
- 分数Nova 2.0 Lite (high)Amazon智能体指数首次评测→3.1
- 分数Nova 2.0 Pro Preview (low)Amazon智能指数19.8→20+1.0%
- 分数Nova 2.0 Pro Preview (low)Amazon智能体指数首次评测→6.4
- 分数Nova 2.0 Pro Preview (medium)Amazon智能指数22.1→22-0.5%
- 分数Nova 2.0 Pro Preview (medium)Amazon智能体指数首次评测→7
- 分数Nova 2.0 Pro Preview (Non-reasoning)Amazon智能体指数首次评测→2.9
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能指数8.6→8.8+2.3%
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能体指数首次评测→2.3
- 分数Phi-4 Mini InstructMicrosoft智能指数5.7→6.2+8.8%
- 分数Phi-4 Mini InstructMicrosoft智能体指数首次评测→0.3
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能指数2.9→3.5+21%
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能体指数首次评测→0.4
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能指数5.2→5-3.8%
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能体指数首次评测→5.7
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能指数5.3→5.7+7.5%
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能体指数首次评测→0.8
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数7.4→7.5+1.4%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能体指数首次评测→0.8
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能指数24.6→24.5-0.4%
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能体指数首次评测→19.1
- 分数Ring-2.6-1TInclusionAI智能指数31.3→31.1-0.6%
- 分数Ring-2.6-1TInclusionAI智能体指数首次评测→18.9
- 分数Trinity Large ThinkingArcee AI智能指数18.4→18.6+1.1%
- 分数Trinity Large ThinkingArcee AI智能体指数首次评测→3.7
08-07
2026年8月7日692 条变化- 新增Ling 3.0 TinyInclusionAI智能指数 23.9
- 新增Ling 3.0 FlashInclusionAI智能指数 37.8
- 新增Muse Spark 1.2 (xhigh)Meta智能指数 56.8
- 新增Qwen3.8 MaxAlibaba智能指数 58.1
- 价格DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek输入价$0.140→$0.130-7.1%
- 价格GLM-5.2 (max)Z AI输入价$1.35→$1.30-3.7%
- 价格GLM-5.2 (max)Z AI输出价$4.29→$4.18-2.6%
- 价格gpt-oss-120b (high)OpenAI输出价$0.600→$0.590-1.7%
- 价格gpt-oss-20b (high)OpenAI输入价$0.070→$0.060-14%
- 价格gpt-oss-20b (high)OpenAI输出价$0.200→$0.190-5.0%
- 价格gpt-oss-20b (low)OpenAI输出价$0.230→$0.200-13%
- 价格Llama 3.1 Instruct 405BMeta输入价$2.50→指标撤下
- 价格Llama 3.1 Instruct 405BMeta输出价$10.00→指标撤下
- 价格Llama 3.1 Instruct 8BMeta输入价$0.080→$0.050-38%
- 价格Llama 3.1 Instruct 8BMeta输出价$0.090→$0.080-11%
- 价格Llama 3.3 Instruct 70BMeta输入价$0.600→$0.660+10%
- 价格Llama 4 MaverickMeta输出价$0.910→$0.850-6.6%
- 分数Agnes 2.5 Pro AlphaSapiens AI智能指数38.8→39.7+2.3%
- 分数Agnes 2.5 Pro AlphaSapiens AI智能体指数25.2→25.4+0.8%
- 分数Apertus 70B InstructSwiss AI Initiative智能指数2.4→2-17%
- 分数Apriel-v1.5-15B-ThinkerServiceNow智能指数21.2→21.6+1.9%
- 分数Apriel-v1.6-15B-ThinkerServiceNow智能指数20.5→20.8+1.5%
- 分数Arctic InstructSnowflake智能指数3.4→3-12%
- 分数Celeris-1Celeris智能指数11.8→12.4+5.1%
- 分数Celeris-1Celeris智能体指数2.8→2.4-14%
- 分数Claude 2.0Anthropic智能指数3.6→3.3-8.3%
- 分数Claude 2.1Anthropic智能指数3.9→3.5-10%
- 分数Claude 3 HaikuAnthropic智能指数3.9→3.5-10%
- 分数Claude 3 SonnetAnthropic智能指数4.7→4.4-6.4%
- 分数Claude 3.5 HaikuAnthropic智能指数12.3→12.2-0.8%
- 分数Claude 3.5 HaikuAnthropic智能体指数1.9→指标撤下
- 分数Claude 3.5 Sonnet (June '24)Anthropic智能指数8.3→8.1-2.4%
- 分数Claude 3.5 Sonnet (Oct '24)Anthropic智能指数9.9→9.8-1.0%
- 分数Claude 3.7 Sonnet (Non-reasoning)Anthropic智能指数23.5→23.9+1.7%
- 分数Claude 3.7 Sonnet (Reasoning)Anthropic智能指数27.1→27.6+1.8%
- 分数Claude 4 Opus (Non-reasoning)Anthropic智能指数25.5→26+2.0%
- 分数Claude 4 Opus (Reasoning)Anthropic智能指数31→31.7+2.3%
- 分数Claude 4 Sonnet (Non-reasoning)Anthropic智能指数25.5→26+2.0%
- 分数Claude 4 Sonnet (Reasoning)Anthropic智能指数28.9→29.8+3.1%
- 分数Claude 4 Sonnet (Reasoning)Anthropic智能体指数16.6→17.6+6.0%
- 分数Claude 4.1 Opus (Non-reasoning)Anthropic智能指数28.2→28.8+2.1%
- 分数Claude 4.1 Opus (Reasoning)Anthropic智能指数33.7→34.5+2.4%
- 分数Claude 4.5 Haiku (Non-reasoning)Anthropic智能指数23.7→24.1+1.7%
- 分数Claude 4.5 Haiku (Reasoning)Anthropic智能指数29.6→29.9+1.0%
- 分数Claude 4.5 Haiku (Reasoning)Anthropic智能体指数16.4→16.5+0.6%
- 分数Claude 4.5 Sonnet (Non-reasoning)Anthropic智能指数29.3→29.9+2.0%
- 分数Claude 4.5 Sonnet (Reasoning)Anthropic智能指数36.4→37.4+2.7%
- 分数Claude 4.5 Sonnet (Reasoning)Anthropic智能体指数24.6→26.4+7.3%
- 分数Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic智能指数59.9→62.1+3.7%
- 分数Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic智能体指数52.8→56.6+7.2%
- 分数Claude InstantAnthropic智能指数2.1→1.7-19%
- 分数Claude Opus 4.5 (Non-reasoning)Anthropic智能指数34.7→35.6+2.6%
- 分数Claude Opus 4.5 (Reasoning)Anthropic智能指数40.8→41.9+2.7%
- 分数Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能指数43.7→44.9+2.7%
- 分数Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic智能指数37.8→38.8+2.6%
- 分数Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic智能指数53.5→55+2.8%
- 分数Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic智能体指数44.4→46.3+4.3%
- 分数Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic智能指数42.7→43.9+2.8%
- 分数Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能指数55.7→57.3+2.9%
- 分数Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic智能体指数47.2→49.4+4.7%
- 分数Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能指数58.9→61.5+4.4%
- 分数Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能体指数52.1→56.1+7.7%
- 分数Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能指数50.6→52.5+3.8%
- 分数Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能体指数39.8→42.1+5.8%
- 分数Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数60.7→63.1+4.0%
- 分数Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能体指数55.3→59.2+7.1%
- 分数Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能指数56.3→58.6+4.1%
- 分数Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能体指数47.1→50.4+7.0%
- 分数Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能指数60.1→62.5+4.0%
- 分数Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能体指数54.5→58.4+7.2%
- 分数Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能指数47.2→48.4+2.5%
- 分数Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic智能体指数40.8→42.1+3.2%
- 分数Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic智能指数35.9→36.8+2.5%
- 分数Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic智能指数34.3→35.1+2.3%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数53.4→55.3+3.6%
- 分数Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic智能体指数46.7→49.7+6.4%
- 分数Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic智能指数41.7→42.6+2.2%
- 分数Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic智能体指数33.7→34.2+1.5%
- 分数Command ACohere智能指数7.7→7.5-2.6%
- 分数Command A+Cohere智能指数22.5→22.8+1.3%
- 分数Command-R (Mar '24)Cohere智能指数2.1→1.7-19%
- 分数Command-R+ (Apr '24)Cohere智能指数3→2.6-13%
- 分数DBRX InstructDatabricks智能指数3→2.6-13%
- 分数DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research智能指数2.3→1.9-17%
- 分数DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research智能指数5.3→5-5.7%
- 分数DeepSeek Coder V2 Lite InstructDeepSeek智能指数3.1→2.7-13%
- 分数DeepSeek LLM 67B Chat (V1)DeepSeek智能指数3→2.6-13%
- 分数DeepSeek R1 (Jan '25)DeepSeek智能指数18.5→18.6+0.5%
- 分数DeepSeek R1 0528 (May '25)DeepSeek智能指数20.1→20.4+1.5%
- 分数DeepSeek R1 0528 Qwen3 8BDeepSeek智能指数10.4→10.3-1.0%
- 分数DeepSeek R1 Distill Llama 70BDeepSeek智能指数9.9→9.8-1.0%
- 分数DeepSeek R1 Distill Llama 8BDeepSeek智能指数6.4→6.2-3.1%
- 分数DeepSeek R1 Distill Qwen 1.5BDeepSeek智能指数3.7→3.3-11%
- 分数DeepSeek R1 Distill Qwen 14BDeepSeek智能指数9.8→9.7-1.0%
- 分数DeepSeek V3 0324DeepSeek智能指数15.4→15.2-1.3%
- 分数DeepSeek V3 0324DeepSeek智能体指数1.5→1.6+6.7%
- 分数DeepSeek V3.1 (Non-reasoning)DeepSeek智能指数21→21.4+1.9%
- 分数DeepSeek V3.1 (Reasoning)DeepSeek智能指数20.7→21+1.4%
- 分数DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek智能指数21.4→21.7+1.4%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能指数30.4→31.1+2.3%
- 分数DeepSeek V3.1 Terminus (Reasoning)DeepSeek智能体指数18.1→指标撤下
- 分数DeepSeek V3.2 (Non-reasoning)DeepSeek智能指数24.7→25.1+1.6%
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能指数32→32.8+2.5%
- 分数DeepSeek V3.2 (Reasoning)DeepSeek智能体指数18.3→指标撤下
- 分数DeepSeek V3.2 Exp (Non-reasoning)DeepSeek智能指数21.3→21.7+1.9%
- 分数DeepSeek V3.2 Exp (Reasoning)DeepSeek智能指数25.4→25.9+2.0%
- 分数DeepSeek V3.2 SpecialeDeepSeek智能指数22.2→22.6+1.8%
- 分数DeepSeek V4 Flash 0420 (Non-reasoning)DeepSeek智能指数28.7→29.3+2.1%
- 分数DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek智能指数37.5→39+4.0%
- 分数DeepSeek V4 Flash 0420 (Reasoning, High Effort)DeepSeek智能体指数28.2→30.3+7.4%
- 分数DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek智能指数40.3→42.1+4.5%
- 分数DeepSeek V4 Flash 0420 (Reasoning, Max Effort)DeepSeek智能体指数31.1→33.7+8.4%
- 分数DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能指数49.9→51.8+3.8%
- 分数DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能体指数45.7→48.4+5.9%
- 分数DeepSeek V4 Pro 0424 (Non-reasoning)DeepSeek智能指数31.2→31.9+2.2%
- 分数DeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeek智能指数43.1→43.7+1.4%
- 分数DeepSeek V4 Pro 0424 (Reasoning, High Effort)DeepSeek智能体指数34.4→35.3+2.6%
- 分数DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek智能指数44.3→45.3+2.3%
- 分数DeepSeek V4 Pro 0424 (Reasoning, Max Effort)DeepSeek智能体指数36.4→37.8+3.8%
- 分数DeepSeek-Coder-V2DeepSeek智能指数5.1→4.7-7.8%
- 分数DeepSeek-V2-ChatDeepSeek智能指数3.6→3.3-8.3%
- 分数DeepSeek-V2.5DeepSeek智能指数6.6→6.4-3.0%
- 分数DeepSeek-V2.5 (Dec '24)DeepSeek智能指数6.8→6.5-4.4%
- 分数Devstral Small (Jul '25)Mistral智能指数9.3→9.1-2.2%
- 分数Devstral Small 2Mistral智能指数17.4→17.7+1.7%
- 分数Devstral Small 2Mistral智能体指数10.5→10.8+2.9%
- 分数DiffusionGemma 26B A4BGoogle智能体指数2.2→指标撤下
- 分数Doubao Seed CodeByteDance Seed智能指数26→26.5+1.9%
- 分数ERNIE 4.5 300B A47BBaidu智能指数9→8.9-1.1%
- 分数ERNIE 5.0 Thinking PreviewBaidu智能指数21.9→22.3+1.8%
- 分数Exaone 4.0 1.2B (Non-reasoning)LG AI Research智能指数2.8→2.4-14%
- 分数Exaone 4.0 1.2B (Reasoning)LG AI Research智能指数2.9→2.5-14%
- 分数EXAONE 4.0 32B (Non-reasoning)LG AI Research智能指数6→5.7-5.0%
- 分数EXAONE 4.0 32B (Reasoning)LG AI Research智能指数10.6→10.5-0.9%
- 分数EXAONE 4.5 33BLG AI Research智能指数20.2→20.5+1.5%
- 分数EXAONE 4.5 33BLG AI Research智能体指数9.7→指标撤下
- 分数Falcon-H1R-7BTII UAE智能指数9.8→9.7-1.0%
- 分数G9v3-39A5BAI9Stars智能指数30.9→31.6+2.3%
- 分数G9v3-39A5BAI9Stars智能体指数28.7→指标撤下
- 分数G9v3-3BAI9Stars智能指数16.1→16.2+0.6%
- 分数G9v3-3BAI9Stars智能体指数14→指标撤下
- 分数Gemini 1.0 ProGoogle智能指数3.1→2.7-13%
- 分数Gemini 1.0 UltraGoogle智能指数4.6→4.3-6.5%
- 分数Gemini 1.5 Flash (May '24)Google智能指数4.9→4.6-6.1%
- 分数Gemini 1.5 Flash (Sep '24)Google智能指数8→7.8-2.5%
- 分数Gemini 1.5 Flash-8BGoogle智能指数5.5→5.2-5.5%
- 分数Gemini 1.5 Pro (May '24)Google智能指数6.3→6.1-3.2%
- 分数Gemini 1.5 Pro (Sep '24)Google智能指数10→9.9-1.0%
- 分数Gemini 2.0 Flash (experimental)Google智能指数10.7→10.6-0.9%
- 分数Gemini 2.0 Flash (Feb '25)Google智能指数12.3→12.2-0.8%
- 分数Gemini 2.0 Flash Thinking Experimental (Dec '24)Google智能指数6.6→6.4-3.0%
- 分数Gemini 2.0 Flash-Lite (Feb '25)Google智能指数8.8→8.6-2.3%
- 分数Gemini 2.0 Flash-Lite (Preview)Google智能指数8.6→8.4-2.3%
- 分数Gemini 2.5 Flash (Non-reasoning)Google智能指数14.1→14.2+0.7%
- 分数Gemini 2.5 Flash (Reasoning)Google智能指数20.1→20.3+1.0%
- 分数Gemini 2.5 Flash Preview (Non-reasoning)Google智能指数11.7→11.6-0.9%
- 分数Gemini 2.5 Flash Preview (Reasoning)Google智能指数17.5→17.7+1.1%
- 分数Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google智能指数18.8→19.1+1.6%
- 分数Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google智能指数23.8→24.2+1.7%
- 分数Gemini 2.5 Flash-Lite (Non-reasoning)Google智能指数6.9→6.7-2.9%
- 分数Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google智能指数15.1→15.2+0.7%
- 分数Gemini 2.5 ProGoogle智能指数25.8→25.9+0.4%
- 分数Gemini 2.5 ProGoogle智能体指数7.1→7.2+1.4%
- 分数Gemini 2.5 Pro Preview (Mar' 25)Google智能指数23→23.4+1.7%
- 分数Gemini 2.5 Pro Preview (May' 25)Google智能指数22.3→22.7+1.8%
- 分数Gemini 3 Flash Preview (Non-reasoning)Google智能指数27.4→27.9+1.8%
- 分数Gemini 3 Flash Preview (Reasoning)Google智能指数37.8→38.7+2.4%
- 分数Gemini 3 Pro Preview (high)Google智能指数39.6→40.6+2.5%
- 分数Gemini 3 Pro Preview (low)Google智能指数33.1→33.9+2.4%
- 分数Gemini 3.1 Flash-LiteGoogle智能指数25→25.6+2.4%
- 分数Gemini 3.1 Flash-LiteGoogle智能体指数6.2→6.5+4.8%
- 分数Gemini 3.1 Pro PreviewGoogle智能指数46.5→47.7+2.6%
- 分数Gemini 3.1 Pro PreviewGoogle智能体指数21.4→23+7.5%
- 分数Gemini 3.5 Flash (high)Google智能指数50.2→52+3.6%
- 分数Gemini 3.5 Flash (high)Google智能体指数37.4→39.7+6.1%
- 分数Gemini 3.5 Flash (medium)Google智能指数45.4→46.7+2.9%
- 分数Gemini 3.5 Flash (minimal)Google智能指数34.9→35.8+2.6%
- 分数Gemini 3.5 Flash-LiteGoogle智能指数36.5→37.4+2.5%
- 分数Gemini 3.5 Flash-LiteGoogle智能体指数26.8→27.2+1.5%
- 分数Gemini 3.6 Flash (high)Google智能指数50.1→51.6+3.0%
- 分数Gemini 3.6 Flash (high)Google智能体指数38.7→40.5+4.7%
- 分数Gemma 3 270MGoogle智能指数2.4→2-17%
- 分数Gemma 3 4B InstructGoogle智能指数1.1→1-9.1%
- 分数Gemma 3n E4B InstructGoogle智能指数1.2→1-17%
- 分数Gemma 3n E4B Instruct Preview (May '25)Google智能指数4.6→4.2-8.7%
- 分数Gemma 4 12B (Reasoning)Google智能指数21.8→22.2+1.8%
- 分数Gemma 4 12B (Reasoning)Google智能体指数7.9→指标撤下
- 分数Gemma 4 26B A4B (Non-reasoning)Google智能指数20.1→20.4+1.5%
- 分数Gemma 4 26B A4B (Reasoning)Google智能指数25.7→26.1+1.6%
- 分数Gemma 4 31B (Non-reasoning)Google智能指数21.8→22.3+2.3%
- 分数Gemma 4 31B (Non-reasoning)Google智能体指数11→11.1+0.9%
- 分数Gemma 4 31B (Reasoning)Google智能指数29.4→30+2.0%
- 分数Gemma 4 31B (Reasoning)Google智能体指数14.4→指标撤下
- 分数Gemma 4 E2B (Non-reasoning)Google智能指数6.4→6.2-3.1%
- 分数Gemma 4 E2B (Reasoning)Google智能指数9.7→9.5-2.1%
- 分数Gemma 4 E2B (Reasoning)Google智能体指数1.5→指标撤下
- 分数Gemma 4 E4B (Non-reasoning)Google智能指数8.9→8.7-2.2%
- 分数Gemma 4 E4B (Reasoning)Google智能体指数1.8→指标撤下
- 分数GLM 5V Turbo (Reasoning)Z AI智能指数34.5→35.3+2.3%
- 分数GLM-4.5 (Reasoning)Z AI智能指数19.5→19.7+1.0%
- 分数GLM-4.5-AirZ AI智能指数16.5→16.7+1.2%
- 分数GLM-4.5V (Non-reasoning)Z AI智能指数7→6.8-2.9%
- 分数GLM-4.5V (Reasoning)Z AI智能指数9.1→9-1.1%
- 分数GLM-4.6 (Non-reasoning)Z AI智能指数23→23.4+1.7%
- 分数GLM-4.6 (Reasoning)Z AI智能指数28.7→29.3+2.1%
- 分数GLM-4.6 (Reasoning)Z AI智能体指数17.7→18.6+5.1%
- 分数GLM-4.6V (Non-reasoning)Z AI智能指数11→10.9-0.9%
- 分数GLM-4.6V (Reasoning)Z AI智能指数16.8→16.9+0.6%
- 分数GLM-4.7 (Non-reasoning)Z AI智能指数26.6→27.1+1.9%
- 分数GLM-4.7 (Reasoning)Z AI智能指数33.7→34.5+2.4%
- 分数GLM-4.7 (Reasoning)Z AI智能体指数25.4→26.2+3.1%
- 分数GLM-4.7-Flash (Non-reasoning)Z AI智能指数15.5→15.6+0.6%
- 分数GLM-4.7-Flash (Reasoning)Z AI智能指数22.9→23.3+1.7%
- 分数GLM-5 (Non-reasoning)Z AI智能指数32.4→33.2+2.5%
- 分数GLM-5 (Reasoning)Z AI智能指数39.5→40.6+2.8%
- 分数GLM-5-TurboZ AI智能指数38.1→39.1+2.6%
- 分数GLM-5.1 (Non-reasoning)Z AI智能指数35.4→36.3+2.5%
- 分数GLM-5.1 (Reasoning)Z AI智能指数40.2→41+2.0%
- 分数GLM-5.1 (Reasoning)Z AI智能体指数29.9→30.6+2.3%
- 分数GLM-5.2 (max)Z AI智能指数51.1→52.6+2.9%
- 分数GLM-5.2 (max)Z AI智能体指数43.1→45.7+6.0%
- 分数GLM-5.2 (Non-reasoning)Z AI智能指数34.1→34.8+2.1%
- 分数GLM-5.2 (Non-reasoning)Z AI智能体指数34.8→35.1+0.9%
- 分数GPT-3.5 TurboOpenAI智能指数3.6→3.2-11%
- 分数GPT-4OpenAI智能指数7→6.8-2.9%
- 分数GPT-4 TurboOpenAI智能指数7.9→7.7-2.5%
- 分数GPT-4.1OpenAI智能指数19.4→19.6+1.0%
- 分数GPT-4.1 miniOpenAI智能体指数1.7→1.8+5.9%
- 分数GPT-4o (Aug '24)OpenAI智能指数9.6→9.4-2.1%
- 分数GPT-4o (ChatGPT)OpenAI智能指数8.2→8.1-1.2%
- 分数GPT-4o (May '24)OpenAI智能指数8.6→8.4-2.3%
- 分数GPT-4o (Nov '24)OpenAI智能指数11.2→11.1-0.9%
- 分数GPT-4o miniOpenAI智能指数6.9→6.7-2.9%
- 分数GPT-5 (ChatGPT)OpenAI智能指数15.3→15.4+0.7%
- 分数GPT-5 (high)OpenAI智能指数34.7→35.3+1.7%
- 分数GPT-5 (high)OpenAI智能体指数25.7→26.5+3.1%
- 分数GPT-5 (low)OpenAI智能指数31.2→31.9+2.2%
- 分数GPT-5 (medium)OpenAI智能指数33.7→34.6+2.7%
- 分数GPT-5 (minimal)OpenAI智能指数17.2→17.3+0.6%
- 分数GPT-5 Codex (high)OpenAI智能指数36.1→37+2.5%
- 分数GPT-5 mini (high)OpenAI智能指数25.3→25.8+2.0%
- 分数GPT-5 mini (high)OpenAI智能体指数19.4→19.6+1.0%
- 分数GPT-5 mini (medium)OpenAI智能指数30.9→31.6+2.3%
- 分数GPT-5 nano (high)OpenAI智能指数19.9→20.1+1.0%
- 分数GPT-5 nano (medium)OpenAI智能指数19→19.2+1.1%
- 分数GPT-5 nano (minimal)OpenAI智能指数8→7.8-2.5%
- 分数GPT-5.1 (high)OpenAI智能指数36.9→37.5+1.6%
- 分数GPT-5.1 (high)OpenAI智能体指数21→21.6+2.9%
- 分数GPT-5.1 (Non-reasoning)OpenAI智能指数20.4→20.7+1.5%
- 分数GPT-5.1 Codex (high)OpenAI智能指数34.7→35.6+2.6%
- 分数GPT-5.1 Codex mini (high)OpenAI智能指数30.6→31.3+2.3%
- 分数GPT-5.2 (medium)OpenAI智能指数38→38.9+2.4%
- 分数GPT-5.2 (Non-reasoning)OpenAI智能指数26→26.5+1.9%
- 分数GPT-5.2 (xhigh)OpenAI智能指数42.2→43.3+2.6%
- 分数GPT-5.2 Codex (xhigh)OpenAI智能指数40.1→41.2+2.7%
- 分数GPT-5.3 Codex (xhigh)OpenAI智能指数44.3→45.5+2.7%
- 分数GPT-5.4 (low)OpenAI智能指数39.1→40.2+2.8%
- 分数GPT-5.4 (Non-reasoning)OpenAI智能指数27.7→28.3+2.2%
- 分数GPT-5.4 (xhigh)OpenAI智能指数51.4→53.1+3.3%
- 分数GPT-5.4 (xhigh)OpenAI智能体指数41.1→44.2+7.5%
- 分数GPT-5.4 mini (medium)OpenAI智能指数29.8→30.5+2.3%
- 分数GPT-5.4 mini (Non-Reasoning)OpenAI智能指数16.6→16.8+1.2%
- 分数GPT-5.4 mini (xhigh)OpenAI智能指数40→40.9+2.2%
- 分数GPT-5.4 mini (xhigh)OpenAI智能体指数30.2→31.5+4.3%
- 分数GPT-5.4 nano (medium)OpenAI智能指数30.2→30.8+2.0%
- 分数GPT-5.4 nano (Non-Reasoning)OpenAI智能指数17.6→17.8+1.1%
- 分数GPT-5.4 nano (xhigh)OpenAI智能指数38.2→39.7+3.9%
- 分数GPT-5.4 nano (xhigh)OpenAI智能体指数27.5→29.7+8.0%
- 分数GPT-5.5 (high)OpenAI智能指数53.1→54.7+3.0%
- 分数GPT-5.5 (high)OpenAI智能体指数43.5→45.9+5.5%
- 分数GPT-5.5 (low)OpenAI智能指数43.5→44.5+2.3%
- 分数GPT-5.5 (low)OpenAI智能体指数30.4→31.7+4.3%
- 分数GPT-5.5 (medium)OpenAI智能指数50.4→51.4+2.0%
- 分数GPT-5.5 (medium)OpenAI智能体指数37.8→39.2+3.7%
- 分数GPT-5.5 (Non-reasoning)OpenAI智能指数35.4→35.8+1.1%
- 分数GPT-5.5 (Non-reasoning)OpenAI智能体指数25.8→26.1+1.2%
- 分数GPT-5.5 (xhigh)OpenAI智能指数54.8→56.3+2.7%
- 分数GPT-5.5 (xhigh)OpenAI智能体指数44.9→47.4+5.6%
- 分数GPT-5.5 Instant (June 2026)OpenAI智能指数28.9→29.2+1.0%
- 分数GPT-5.5 Instant (June 2026)OpenAI智能体指数11→11.1+0.9%
- 分数GPT-5.5 Instant (May 2026)OpenAI智能指数33.5→34.3+2.4%
- 分数GPT-5.6 Luna (high)OpenAI智能指数46.1→47+2.0%
- 分数GPT-5.6 Luna (high)OpenAI智能体指数40.1→41+2.2%
- 分数GPT-5.6 Luna (low)OpenAI智能指数33.3→33.9+1.8%
- 分数GPT-5.6 Luna (low)OpenAI智能体指数25.4→25.7+1.2%
- 分数GPT-5.6 Luna (max)OpenAI智能指数51.2→52.3+2.1%
- 分数GPT-5.6 Luna (max)OpenAI智能体指数45.6→46.9+2.9%
- 分数GPT-5.6 Luna (medium)OpenAI智能指数38.1→38.9+2.1%
- 分数GPT-5.6 Luna (medium)OpenAI智能体指数31→31.8+2.6%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能指数26.6→26.8+0.8%
- 分数GPT-5.6 Luna (Non-reasoning)OpenAI智能体指数22→22.2+0.9%
- 分数GPT-5.6 Luna (xhigh)OpenAI智能指数49.1→50.1+2.0%
- 分数GPT-5.6 Luna (xhigh)OpenAI智能体指数42.9→44.4+3.5%
- 分数GPT-5.6 Sol (high)OpenAI智能指数55.9→57.3+2.5%
- 分数GPT-5.6 Sol (high)OpenAI智能体指数48.5→50.6+4.3%
- 分数GPT-5.6 Sol (low)OpenAI智能指数49.4→50.7+2.6%
- 分数GPT-5.6 Sol (low)OpenAI智能体指数40→41.6+4.0%
- 分数GPT-5.6 Sol (max)OpenAI智能指数58.9→60.9+3.4%
- 分数GPT-5.6 Sol (max)OpenAI智能体指数54→57.8+7.0%
- 分数GPT-5.6 Sol (medium)OpenAI智能指数53.6→55.6+3.7%
- 分数GPT-5.6 Sol (medium)OpenAI智能体指数44.5→47.9+7.6%
- 分数GPT-5.6 Sol (Non-reasoning)OpenAI智能指数41.2→41.9+1.7%
- 分数GPT-5.6 Sol (Non-reasoning)OpenAI智能体指数34.9→36+3.2%
- 分数GPT-5.6 Sol (xhigh)OpenAI智能指数57.7→59+2.3%
- 分数GPT-5.6 Sol (xhigh)OpenAI智能体指数51.8→53.6+3.5%
- 分数GPT-5.6 Terra (high)OpenAI智能指数49→50.1+2.2%
- 分数GPT-5.6 Terra (high)OpenAI智能体指数41.3→43.5+5.3%
- 分数GPT-5.6 Terra (low)OpenAI智能指数40.5→41.3+2.0%
- 分数GPT-5.6 Terra (low)OpenAI智能体指数30.6→31.5+2.9%
- 分数GPT-5.6 Terra (max)OpenAI智能指数55→56.6+2.9%
- 分数GPT-5.6 Terra (max)OpenAI智能体指数47.4→50.2+5.9%
- 分数GPT-5.6 Terra (medium)OpenAI智能指数45.6→46.8+2.6%
- 分数GPT-5.6 Terra (medium)OpenAI智能体指数37→39.1+5.7%
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能指数34→34.6+1.8%
- 分数GPT-5.6 Terra (Non-reasoning)OpenAI智能体指数29.3→30.1+2.7%
- 分数GPT-5.6 Terra (xhigh)OpenAI智能指数51.6→52.8+2.3%
- 分数GPT-5.6 Terra (xhigh)OpenAI智能体指数44.7→46.5+4.0%
- 分数gpt-oss-120b (high)OpenAI智能指数23.8→24.1+1.3%
- 分数gpt-oss-120b (high)OpenAI智能体指数13.2→13.4+1.5%
- 分数gpt-oss-20b (high)OpenAI智能指数14.9→15.2+2.0%
- 分数gpt-oss-20b (low)OpenAI智能指数14.3→14.4+0.7%
- 分数Granite 3.3 8B (Non-reasoning)IBM智能指数1.8→1.3-28%
- 分数Granite 4.0 1BIBM智能指数2.1→1.6-24%
- 分数Granite 4.0 H 1BIBM智能指数2.7→2.2-19%
- 分数Granite 4.0 H SmallIBM智能指数5.2→4.9-5.8%
- 分数Granite 4.0 MicroIBM智能指数2.4→2-17%
- 分数Granite 4.1 30BIBM智能指数8.9→8.7-2.2%
- 分数Granite 4.1 30BIBM智能体指数1.3→指标撤下
- 分数Granite 4.1 3BIBM智能指数4.7→4.4-6.4%
- 分数Granite 4.1 3BIBM智能体指数0.4→指标撤下
- 分数Granite 4.1 8BIBM智能指数6.7→6.4-4.5%
- 分数Grok 2 (Dec '24)SpaceXAI智能指数8→7.8-2.5%
- 分数Grok 3SpaceXAI智能指数18.4→18.6+1.1%
- 分数Grok 3 mini Reasoning (high)SpaceXAI智能指数22.5→22.9+1.8%
- 分数Grok 3 Reasoning BetaSpaceXAI智能指数15.1→15.2+0.7%
- 分数Grok 4SpaceXAI智能指数33.3→34.1+2.4%
- 分数Grok 4 Fast (Non-reasoning)SpaceXAI智能指数16.5→16.6+0.6%
- 分数Grok 4 Fast (Reasoning)SpaceXAI智能指数27.4→27.9+1.8%
- 分数Grok 4.1 Fast (Non-reasoning)SpaceXAI智能指数16.9→17+0.6%
- 分数Grok 4.1 Fast (Reasoning)SpaceXAI智能指数30.6→31.3+2.3%
- 分数Grok 4.20 0309 (Non-reasoning)SpaceXAI智能指数22.5→22.9+1.8%
- 分数Grok 4.20 0309 (Reasoning)SpaceXAI智能指数36.5→37.4+2.5%
- 分数Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI智能指数21.8→22.2+1.8%
- 分数Grok 4.20 0309 v2 (Reasoning)SpaceXAI智能指数37→38+2.7%
- 分数Grok 4.3 (high)SpaceXAI智能指数37.6→37.9+0.8%
- 分数Grok 4.3 (high)SpaceXAI智能体指数24.1→24.2+0.4%
- 分数Grok 4.3 (low)SpaceXAI智能指数35.4→36.3+2.5%
- 分数Grok 4.3 (medium)SpaceXAI智能指数36→36.9+2.5%
- 分数Grok 4.3 (Non-reasoning)SpaceXAI智能指数24.8→25+0.8%
- 分数Grok 4.3 (Non-reasoning)SpaceXAI智能体指数22.8→23+0.9%
- 分数Grok 4.5 (high)SpaceXAI智能指数53.8→55.8+3.7%
- 分数Grok 4.5 (high)SpaceXAI智能体指数45.7→48.9+7.0%
- 分数Grok BetaSpaceXAI智能指数7.5→7.3-2.7%
- 分数Grok Build 0.1 0616SpaceXAI智能指数39.8→40.7+2.3%
- 分数Grok Build 0.1 0616SpaceXAI智能体指数28→28.9+3.2%
- 分数Grok Code Fast 1SpaceXAI智能指数21.6→22+1.9%
- 分数Grok-1SpaceXAI智能指数6→5.8-3.3%
- 分数Hermes 3 - Llama-3.1 70BNous Research智能指数5.1→4.8-5.9%
- 分数Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research智能指数8.8→8.6-2.3%
- 分数Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research智能指数9→8.8-2.2%
- 分数Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research智能指数6.9→6.7-2.9%
- 分数Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research智能指数10→9.9-1.0%
- 分数Hy3Tencent智能指数41.2→42.2+2.4%
- 分数Hy3Tencent智能体指数30.7→31.4+2.3%
- 分数Hy3-preview (Non-reasoning)Tencent智能指数26.1→26.6+1.9%
- 分数Hy3-preview (Reasoning)Tencent智能指数33.6→34.4+2.4%
- 分数HyperCLOVA X SEED Think (32B)Naver智能指数17→17.2+1.2%
- 分数HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing智能指数17.8→18.3+2.8%
- 分数HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing智能体指数6.7→6.8+1.5%
- 分数Inkling (xhigh)Thinking Machines智能指数40.7→42.3+3.9%
- 分数Inkling (xhigh)Thinking Machines智能体指数32.3→34.1+5.6%
- 分数Inkling SmallThinking Machines智能指数40.2→41.2+2.5%
- 分数Inkling SmallThinking Machines智能体指数30.8→31.9+3.6%
- 分数INTELLECT-3 (based on GLM-4.5-Air)Prime Intellect智能指数15.6→15.7+0.6%
- 分数Jamba 1.5 LargeAI21 Labs智能指数5.1→4.8-5.9%
- 分数Jamba 1.5 MiniAI21 Labs智能指数2.7→2.3-15%
- 分数Jamba 1.6 LargeAI21 Labs智能指数5→4.7-6.0%
- 分数Jamba 1.6 MiniAI21 Labs智能指数2.6→2.1-19%
- 分数Jamba 1.7 LargeAI21 Labs智能指数5.3→5-5.7%
- 分数Jamba 1.7 MiniAI21 Labs智能指数2.7→2.3-15%
- 分数Jamba Reasoning 3BAI21 Labs智能指数4.1→3.8-7.3%
- 分数JT-35B-FlashChina Mobile智能指数28.4→29+2.1%
- 分数JT-4.1 Flash 236B A21BChina Mobile智能指数38.8→39.9+2.8%
- 分数JT-4.1 Flash 236B A21BChina Mobile智能体指数34.9→指标撤下
- 分数JT-MINIChina Mobile智能指数18.5→18.8+1.6%
- 分数K-EXAONE (Non-reasoning)LG AI Research智能指数16.7→16.9+1.2%
- 分数K-EXAONE (Reasoning)LG AI Research智能指数22.1→22.5+1.8%
- 分数K-EXAONE (Reasoning)LG AI Research智能体指数8→指标撤下
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能指数17.3→17.4+0.6%
- 分数K2 Think V2MBZUAI Institute of Foundation Models智能体指数1.8→指标撤下
- 分数K2-V2 (low)MBZUAI Institute of Foundation Models智能指数8.6→8.4-2.3%
- 分数KAT Coder Pro V2KwaiKAT智能指数33.7→34.5+2.4%
- 分数KAT Coder Pro V2KwaiKAT智能体指数15.5→指标撤下
- 分数KAT-Coder-Pro V1KwaiKAT智能指数28.3→28.9+2.1%
- 分数Kimi K2Kimi智能指数19.4→19.7+1.5%
- 分数Kimi K2 0905Kimi智能指数23.5→24+2.1%
- 分数Kimi K2 ThinkingKimi智能指数32.7→33.5+2.4%
- 分数Kimi K2.5 (Non-reasoning)Kimi智能指数29.4→30.1+2.4%
- 分数Kimi K2.5 (Reasoning)Kimi智能指数35.4→36+1.7%
- 分数Kimi K2.6Kimi智能指数44.2→45.1+2.0%
- 分数Kimi K2.6Kimi智能体指数30.3→31.2+3.0%
- 分数Kimi K2.6 (Non-reasoning)Kimi智能指数34.6→35.4+2.3%
- 分数Kimi K2.7 CodeKimi智能指数41.9→43+2.6%
- 分数Kimi K2.7 CodeKimi智能体指数29.6→30.3+2.4%
- 分数Kimi K3 (low)Kimi智能指数46.6→48.3+3.6%
- 分数Kimi K3 (low)Kimi智能体指数37→39.6+7.0%
- 分数Kimi K3 (max)Kimi智能指数57.1→59.7+4.6%
- 分数Kimi K3 (max)Kimi智能体指数50.1→54.3+8.4%
- 分数Kimi Linear 48B A3B InstructKimi智能指数8.5→8.4-1.2%
- 分数LFM 40BLiquid AI智能指数3.4→3-12%
- 分数LFM2 1.2BLiquid AI智能指数1.1→1-9.1%
- 分数LFM2 2.6BLiquid AI智能指数2.7→2.3-15%
- 分数LFM2 24B A2BLiquid AI智能指数4.9→4.6-6.1%
- 分数LFM2 8B A1BLiquid AI智能指数1.8→1.3-28%
- 分数LFM2.5-1.2B-InstructLiquid AI智能指数2.7→2.3-15%
- 分数LFM2.5-1.2B-ThinkingLiquid AI智能指数2.7→2.3-15%
- 分数LFM2.5-8B-A1BLiquid AI智能指数8.3→8.1-2.4%
- 分数Ling 2.6 FlashInclusionAI智能指数14.1→14.2+0.7%
- 分数Ling-1TInclusionAI智能指数12.8→12.7-0.8%
- 分数Ling-2.6-1TInclusionAI智能指数26.1→26.6+1.9%
- 分数Ling-flash-2.0InclusionAI智能指数9.7→9.6-1.0%
- 分数Ling-mini-2.0InclusionAI智能指数3.8→3.4-11%
- 分数Llama 2 Chat 13BMeta智能指数3→2.6-13%
- 分数Llama 2 Chat 70BMeta智能指数3→2.6-13%
- 分数Llama 2 Chat 7BMeta智能指数4.3→3.9-9.3%
- 分数Llama 3 Instruct 70BMeta智能指数3.5→3.1-11%
- 分数Llama 3 Instruct 8BMeta智能指数1.2→1-17%
- 分数Llama 3.1 Instruct 405BMeta智能指数8.5→8.3-2.4%
- 分数Llama 3.1 Instruct 70BMeta智能指数6.8→6.5-4.4%
- 分数Llama 3.1 Instruct 8BMeta智能指数7.6→7.4-2.6%
- 分数Llama 3.1 Instruct 8BMeta智能体指数0.5→指标撤下
- 分数Llama 3.1 Nemotron Instruct 70BNVIDIA智能指数7.6→7.4-2.6%
- 分数Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIA智能指数8.5→8.4-1.2%
- 分数Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA智能指数9.1→8.9-2.2%
- 分数Llama 3.1 Tulu3 405BAllen Institute for AI智能指数8.3→8.1-2.4%
- 分数Llama 3.2 Instruct 11B (Vision)Meta智能指数3.3→3-9.1%
- 分数Llama 3.2 Instruct 1BMeta智能指数1.1→1-9.1%
- 分数Llama 3.2 Instruct 3BMeta智能指数4.2→3.9-7.1%
- 分数Llama 3.2 Instruct 90B (Vision)Meta智能指数6.2→6-3.2%
- 分数Llama 3.3 Instruct 70BMeta智能指数9.4→9.3-1.1%
- 分数Llama 3.3 Instruct 70BMeta智能体指数0.3→指标撤下
- 分数Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA智能指数8.5→8.3-2.4%
- 分数Llama 4 MaverickMeta智能指数14.3→14.5+1.4%
- 分数Llama 4 MaverickMeta智能体指数1.3→1.2-7.7%
- 分数Llama 4 ScoutMeta智能指数10→10.3+3.0%
- 分数Llama 65BMeta智能指数2.1→1.7-19%
- 分数Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA智能指数8.7→8.5-2.3%
- 分数LongCat 2.0LongCat智能指数33.5→34.3+2.4%
- 分数LongCat 2.0LongCat智能体指数21.8→指标撤下
- 分数LongCat Flash LiteLongCat智能指数17.2→17.4+1.2%
- 分数Magistral Medium 1.2Mistral智能指数17.9→18+0.6%
- 分数Magistral Small 1Mistral智能指数10.7→10.6-0.9%
- 分数Magistral Small 1.2Mistral智能指数11.3→11.5+1.8%
- 分数Magistral Small 1.2Mistral智能体指数1.4→1.5+7.1%
- 分数Mercury 2Inception智能指数21.4→21.9+2.3%
- 分数Mercury 2Inception智能体指数9.6→9.5-1.0%
- 分数Mi:dm K 2.5 ProKorea Telecom智能指数16.4→16.6+1.2%
- 分数MiMo-V2-Flash (Feb 2026)Xiaomi智能指数33.2→34+2.4%
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能指数24.7→25.1+1.6%
- 分数MiMo-V2-Flash (Non-reasoning)Xiaomi智能体指数12→指标撤下
- 分数MiMo-V2-Flash (Reasoning)Xiaomi智能指数31.2→31.9+2.2%
- 分数MiMo-V2-OmniXiaomi智能指数35→35.9+2.6%
- 分数MiMo-V2-Omni-0327Xiaomi智能指数36.4→37.3+2.5%
- 分数MiMo-V2-ProXiaomi智能指数40.3→41.4+2.7%
- 分数MiMo-V2.5Xiaomi智能指数37.2→38+2.2%
- 分数MiMo-V2.5Xiaomi智能体指数23.7→24.4+3.0%
- 分数MiMo-V2.5-ProXiaomi智能指数42.2→42.9+1.7%
- 分数MiMo-V2.5-ProXiaomi智能体指数29.1→29.5+1.4%
- 分数MiMo-V2.5-Pro (Non-reasoning)Xiaomi智能指数27.9→28.4+1.8%
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能指数4.2→3.8-9.5%
- 分数MiniCPM-V 4.6 1.3BOpenBMB智能体指数1.4→指标撤下
- 分数MiniCPM5-1B (Reasoning)OpenBMB智能指数12→11.9-0.8%
- 分数MiniMax M1 40kMiniMax智能指数14.4→14.5+0.7%
- 分数MiniMax M1 80kMiniMax智能指数17.7→17.9+1.1%
- 分数MiniMax-M2MiniMax智能指数28.3→28.9+2.1%
- 分数MiniMax-M2.1MiniMax智能指数31.4→32.1+2.2%
- 分数MiniMax-M2.5MiniMax智能指数33.7→34.5+2.4%
- 分数MiniMax-M2.7MiniMax智能指数38.1→38.9+2.1%
- 分数MiniMax-M2.7MiniMax智能体指数25.6→25.9+1.2%
- 分数MiniMax-M3MiniMax智能指数44.4→45.4+2.3%
- 分数MiniMax-M3MiniMax智能体指数35.4→36.1+2.0%
- 分数Ministral 3 14BMistral智能指数11.1→11.2+0.9%
- 分数Ministral 3 3BMistral智能指数6.8→7.1+4.4%
- 分数Mistral 7B InstructMistral智能指数2.1→1.7-19%
- 分数Mistral Large (Feb '24)Mistral智能指数4.4→4.1-6.8%
- 分数Mistral Large 2 (Jul '24)Mistral智能指数7.3→7-4.1%
- 分数Mistral Large 2 (Nov '24)Mistral智能指数9.1→9-1.1%
- 分数Mistral MediumMistral智能指数3.6→3.2-11%
- 分数Mistral Medium 3.1Mistral智能体指数6.2→6.1-1.6%
- 分数Mistral Medium 3.5Mistral智能指数29.9→30.4+1.7%
- 分数Mistral Medium 3.5Mistral智能体指数19→19.2+1.1%
- 分数Mistral SabaMistral智能指数6.4→6.2-3.1%
- 分数Mistral Small (Feb '24)Mistral智能指数3.6→3.2-11%
- 分数Mistral Small (Sep '24)Mistral智能指数4.7→4.3-8.5%
- 分数Mistral Small 3Mistral智能指数6.9→6.7-2.9%
- 分数Mistral Small 3.1Mistral智能指数14.7→14.9+1.4%
- 分数Mistral Small 3.1Mistral智能体指数5.2→5.3+1.9%
- 分数Mistral Small 3.2Mistral智能指数10.6→10.7+0.9%
- 分数Mistral Small 3.2Mistral智能体指数2→2.1+5.0%
- 分数Mistral Small 4 (Non-reasoning)Mistral智能指数12.4→12.3-0.8%
- 分数Mistral Small 4 (Reasoning)Mistral智能指数19.6→19.7+0.5%
- 分数Mistral Small 4 (Reasoning)Mistral智能体指数4.7→4.6-2.1%
- 分数Mixtral 8x22B InstructMistral智能指数4.4→4-9.1%
- 分数Mixtral 8x7B InstructMistral智能指数2.4→2-17%
- 分数Molmo 7B-DAllen Institute for AI智能指数3.8→3.4-11%
- 分数Molmo2-8BAllen Institute for AI智能指数2→1.6-20%
- 分数Motif 3 (Beta)Motif Technologies智能指数44.1→45.3+2.7%
- 分数Motif 3 (Beta)Motif Technologies智能体指数34.9→指标撤下
- 分数Muse SparkMeta智能指数43.1→44.3+2.8%
- 分数Muse SparkMeta智能体指数28.7→指标撤下
- 分数Muse Spark 1.1 (xhigh)Meta智能指数50.6→53.2+5.1%
- 分数Muse Spark 1.1 (xhigh)Meta智能体指数37.5→39.7+5.9%
- 分数Nanbeige4.1-3BNanbeige智能指数11.1→11-0.9%
- 分数Nanbeige4.1-3BNanbeige智能体指数0→指标撤下
- 分数Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA智能指数14.9→15+0.7%
- 分数Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA智能指数37.8→38.3+1.3%
- 分数Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA智能体指数27.4→27.5+0.4%
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能指数17.6→17.8+1.1%
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能体指数3.7→指标撤下
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能指数41→42.1+2.7%
- 分数Nex-N2-Pro (based on Qwen3.5-397B-A17B)Nex AGI智能体指数31→指标撤下
- 分数North Mini CodeCohere智能指数19.8→20.2+2.0%
- 分数Nova 2.0 Lite (high)Amazon智能指数18.2→20.8+14%
- 分数Nova 2.0 Lite (high)Amazon智能体指数3.1→指标撤下
- 分数Nova 2.0 Lite (low)Amazon智能指数17.8→18+1.1%
- 分数Nova 2.0 Lite (medium)Amazon智能指数19→19.2+1.1%
- 分数Nova 2.0 Omni (low)Amazon智能指数16.6→16.7+0.6%
- 分数Nova 2.0 Omni (medium)Amazon智能指数20.9→21.3+1.9%
- 分数Nova 2.0 Omni (Non-reasoning)Amazon智能指数10.5→10.4-1.0%
- 分数Nova 2.0 Pro Preview (low)Amazon智能指数19.6→19.8+1.0%
- 分数Nova 2.0 Pro Preview (low)Amazon智能体指数6.4→指标撤下
- 分数Nova 2.0 Pro Preview (medium)Amazon智能指数21.8→22.1+1.4%
- 分数Nova 2.0 Pro Preview (medium)Amazon智能体指数7→指标撤下
- 分数Nova 2.0 Pro Preview (Non-reasoning)Amazon智能体指数2.9→指标撤下
- 分数Nova LiteAmazon智能指数6.9→6.7-2.9%
- 分数Nova MicroAmazon智能指数4.7→4.4-6.4%
- 分数Nova ProAmazon智能指数7.7→7.5-2.6%
- 分数NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA智能指数7.4→7.2-2.7%
- 分数NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA智能指数14.2→14.5+2.1%
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能指数8.7→8.6-1.1%
- 分数NVIDIA Nemotron 3 Nano 4BNVIDIA智能体指数2.3→指标撤下
- 分数Nemotron 3 Super 120B A12B (Reasoning)NVIDIA智能指数25.4→25.7+1.2%
- 分数Nemotron 3 Super 120B A12B (Reasoning)NVIDIA智能体指数8.7→8.8+1.1%
- 分数NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA智能指数4.6→4.2-8.7%
- 分数NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA智能指数9→8.8-2.2%
- 分数NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA智能指数7.4→7.2-2.7%
- 分数NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA智能指数8.8→8.7-1.1%
- 分数o1OpenAI智能指数23.4→23.9+2.1%
- 分数o1-previewOpenAI智能指数17→17.2+1.2%
- 分数o1-proOpenAI智能指数18.9→19.1+1.1%
- 分数o3OpenAI智能指数30.4→31.1+2.3%
- 分数o3-miniOpenAI智能指数19→19.2+1.1%
- 分数o3-mini (high)OpenAI智能指数15.6→15.7+0.6%
- 分数o3-proOpenAI智能指数32.5→33.3+2.5%
- 分数o4-mini (high)OpenAI智能指数25.6→26.1+2.0%
- 分数OLMo 2 32BAllen Institute for AI智能指数5→4.7-6.0%
- 分数OLMo 2 7BAllen Institute for AI智能指数3.9→3.5-10%
- 分数Olmo 3 32B ThinkAllen Institute for AI智能指数6.4→6.1-4.7%
- 分数Olmo 3 7B InstructAllen Institute for AI智能指数2.8→2.4-14%
- 分数Olmo 3 7B ThinkAllen Institute for AI智能指数4→3.6-10.0%
- 分数Olmo 3.1 32B InstructAllen Institute for AI智能指数6.5→6.2-4.6%
- 分数Olmo 3.1 32B ThinkAllen Institute for AI智能指数8.1→7.9-2.5%
- 分数OpenChat 3.5 (1210)OpenChat智能指数3→2.6-13%
- 分数PALM-2Google智能指数3.2→2.8-13%
- 分数Phi-3 Mini Instruct 3.8BMicrosoft智能指数4.6→4.3-6.5%
- 分数Phi-4Microsoft智能指数4.9→4.6-6.1%
- 分数Phi-4 Mini InstructMicrosoft智能指数6→5.7-5.0%
- 分数Phi-4 Mini InstructMicrosoft智能体指数0.3→指标撤下
- 分数Phi-4 Multimodal InstructMicrosoft智能指数4.5→4.2-6.7%
- 分数Pixtral LargeMistral智能指数8.1→8-1.2%
- 分数Qwen Chat 14BAlibaba智能指数2.1→1.7-19%
- 分数Qwen Chat 72BAlibaba智能指数3.4→3-12%
- 分数Qwen1.5 Chat 110BAlibaba智能指数4.1→3.7-9.8%
- 分数Qwen2 Instruct 72BAlibaba智能指数6→5.7-5.0%
- 分数Qwen2.5 Coder Instruct 32BAlibaba智能指数7.1→6.9-2.8%
- 分数Qwen2.5 Coder Instruct 7BAlibaba智能指数4.5→4.1-8.9%
- 分数Qwen2.5 Instruct 32BAlibaba智能指数7.5→7.2-4.0%
- 分数Qwen2.5 Instruct 72BAlibaba智能指数9.6→9.4-2.1%
- 分数Qwen2.5 MaxAlibaba智能指数10.2→10.1-1.0%
- 分数Qwen2.5 TurboAlibaba智能指数6.3→6-4.8%
- 分数Qwen3 0.6B (Reasoning)Alibaba智能指数1.3→1-23%
- 分数Qwen3 1.7B (Non-reasoning)Alibaba智能指数1.5→1.1-27%
- 分数Qwen3 1.7B (Reasoning)Alibaba智能指数2.6→2.2-15%
- 分数Qwen3 14B (Non-reasoning)Alibaba智能指数7→6.8-2.9%
- 分数Qwen3 14B (Reasoning)Alibaba智能体指数1.8→1.9+5.6%
- 分数Qwen3 235B A22B (Non-reasoning)Alibaba智能指数10.9→10.8-0.9%
- 分数Qwen3 235B A22B (Reasoning)Alibaba智能指数13.4→13.5+0.7%
- 分数Qwen3 235B A22B 2507 (Reasoning)Alibaba智能指数19.6→19.9+1.5%
- 分数Qwen3 235B A22B 2507 InstructAlibaba智能指数18.2→18.4+1.1%
- 分数Qwen3 30B A3B (Non-reasoning)Alibaba智能指数6.8→6.6-2.9%
- 分数Qwen3 30B A3B (Reasoning)Alibaba智能指数9.3→9.2-1.1%
- 分数Qwen3 30B A3B 2507 (Reasoning)Alibaba智能指数14.4→14.6+1.4%
- 分数Qwen3 30B A3B 2507 InstructAlibaba智能指数9.1→8.9-2.2%
- 分数Qwen3 32B (Non-reasoning)Alibaba智能指数8.6→8.5-1.2%
- 分数Qwen3 32B (Reasoning)Alibaba智能指数11.5→11.4-0.9%
- 分数Qwen3 4B (Non-reasoning)Alibaba智能指数6.8→6.5-4.4%
- 分数Qwen3 4B (Reasoning)Alibaba智能指数8.4→8.2-2.4%
- 分数Qwen3 4B 2507 (Reasoning)Alibaba智能指数12→11.9-0.8%
- 分数Qwen3 4B 2507 InstructAlibaba智能指数7.1→6.9-2.8%
- 分数Qwen3 8B (Non-reasoning)Alibaba智能指数5.1→4.8-5.9%
- 分数Qwen3 8B (Reasoning)Alibaba智能体指数1.5→1.6+6.7%
- 分数Qwen3 Coder 480B A35B InstructAlibaba智能指数18→18.2+1.1%
- 分数Qwen3 Coder NextAlibaba智能指数21.1→21.3+0.9%
- 分数Qwen3 Coder NextAlibaba智能体指数8.8→8.9+1.1%
- 分数Qwen3 MaxAlibaba智能指数24→24.5+2.1%
- 分数Qwen3 Max (Preview)Alibaba智能指数19.2→19.4+1.0%
- 分数Qwen3 Max ThinkingAlibaba智能指数31.7→32.5+2.5%
- 分数Qwen3 Max Thinking (Preview)Alibaba智能指数25→25.5+2.0%
- 分数Qwen3 Next 80B A3B (Reasoning)Alibaba智能指数16.7→16.9+1.2%
- 分数Qwen3 Next 80B A3B InstructAlibaba智能指数13.7→13.8+0.7%
- 分数Qwen3 Omni 30B A3B (Reasoning)Alibaba智能指数9.6→9.5-1.0%
- 分数Qwen3 Omni 30B A3B InstructAlibaba智能指数5.1→4.8-5.9%
- 分数Qwen3 VL 235B A22B (Reasoning)Alibaba智能指数20.6→20.9+1.5%
- 分数Qwen3 VL 235B A22B InstructAlibaba智能指数14.3→14.4+0.7%
- 分数Qwen3 VL 30B A3B (Reasoning)Alibaba智能指数13.3→13.4+0.8%
- 分数Qwen3 VL 30B A3B InstructAlibaba智能指数10→9.9-1.0%
- 分数Qwen3 VL 32B (Reasoning)Alibaba智能指数17.9→18.1+1.1%
- 分数Qwen3 VL 32B InstructAlibaba智能指数11.1→11-0.9%
- 分数Qwen3 VL 4B (Reasoning)Alibaba智能指数7.9→7.7-2.5%
- 分数Qwen3 VL 4B InstructAlibaba智能指数4.1→3.7-9.8%
- 分数Qwen3 VL 8B (Reasoning)Alibaba智能指数10.6→10.5-0.9%
- 分数Qwen3 VL 8B InstructAlibaba智能指数8.4→8.2-2.4%
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能指数3.3→2.9-12%
- 分数Qwen3.5 0.8B (Non-reasoning)Alibaba智能体指数0.4→指标撤下
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能指数5.5→5.2-5.5%
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能体指数5.7→指标撤下
- 分数Qwen3.5 122B A10B (Non-reasoning)Alibaba智能指数27.6→28.2+2.2%
- 分数Qwen3.5 122B A10B (Non-reasoning)Alibaba智能体指数15.8→16.2+2.5%
- 分数Qwen3.5 122B A10B (Reasoning)Alibaba智能指数32.3→32.8+1.5%
- 分数Qwen3.5 122B A10B (Reasoning)Alibaba智能体指数20.7→21.3+2.9%
- 分数Qwen3.5 27B (Non-reasoning)Alibaba智能指数29.3→30+2.4%
- 分数Qwen3.5 27B (Reasoning)Alibaba智能指数33.8→34.6+2.4%
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能指数5.6→5.3-5.4%
- 分数Qwen3.5 2B (Non-reasoning)Alibaba智能体指数0.8→指标撤下
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数7.6→7.4-2.6%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能体指数0.8→指标撤下
- 分数Qwen3.5 35B A3B (Non-reasoning)Alibaba智能指数24→24.3+1.3%
- 分数Qwen3.5 35B A3B (Reasoning)Alibaba智能指数29.3→29.9+2.0%
- 分数Qwen3.5 397B A17B (Non-reasoning)Alibaba智能指数32→32.7+2.2%
- 分数Qwen3.5 397B A17B (Reasoning)Alibaba智能指数33.7→34.3+1.8%
- 分数Qwen3.5 4B (Non-reasoning)Alibaba智能指数16→16.1+0.6%
- 分数Qwen3.5 4B (Reasoning)Alibaba智能指数20.1→20.4+1.5%
- 分数Qwen3.5 9B (Non-reasoning)Alibaba智能指数20.3→20.6+1.5%
- 分数Qwen3.5 9B (Reasoning)Alibaba智能指数21.4→21.8+1.9%
- 分数Qwen3.5 9B (Reasoning)Alibaba智能体指数7.4→7-5.4%
- 分数Qwen3.5 Omni FlashAlibaba智能指数19→19.2+1.1%
- 分数Qwen3.5 Omni PlusAlibaba智能指数30.6→31.3+2.3%
- 分数Qwen3.6 27B (Non-reasoning)Alibaba智能指数30.5→31.3+2.6%
- 分数Qwen3.6 27B (Non-reasoning)Alibaba智能体指数23.3→23.6+1.3%
- 分数Qwen3.6 27B (Reasoning)Alibaba智能指数37.1→37.7+1.6%
- 分数Qwen3.6 27B (Reasoning)Alibaba智能体指数27→27.5+1.9%
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能指数24.2→24.6+1.7%
- 分数Qwen3.6 35B A3B (Non-reasoning)Alibaba智能体指数19.1→指标撤下
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能指数31.6→32.1+1.6%
- 分数Qwen3.6 35B A3B (Reasoning)Alibaba智能体指数21.4→21.6+0.9%
- 分数Qwen3.6 Max PreviewAlibaba智能指数40→41.1+2.8%
- 分数Qwen3.6 PlusAlibaba智能指数39.6→40.5+2.3%
- 分数Qwen3.6 PlusAlibaba智能体指数27.6→29+5.1%
- 分数Qwen3.7 MaxAlibaba智能指数46→46.7+1.5%
- 分数Qwen3.7 MaxAlibaba智能体指数30.6→30.9+1.0%
- 分数Qwen3.7 PlusAlibaba智能指数39→39.4+1.0%
- 分数Qwen3.7 PlusAlibaba智能体指数20.8→20.7-0.5%
- 分数QwQ 32B-PreviewAlibaba智能指数9.2→9.1-1.1%
- 分数R1 1776Perplexity智能指数6.3→6-4.8%
- 分数Reka Flash (Sep '24)Reka AI智能指数6.3→6-4.8%
- 分数Reka Flash 3Reka AI智能指数4.1→3.7-9.8%
- 分数Ring-1TInclusionAI智能指数16.2→16.3+0.6%
- 分数Ring-2.6-1TInclusionAI智能指数30.6→31.3+2.3%
- 分数Ring-2.6-1TInclusionAI智能体指数18.9→指标撤下
- 分数Ring-flash-2.0InclusionAI智能指数8.2→8-2.4%
- 分数Sarvam 30B (high)Sarvam智能指数6.6→6.4-3.0%
- 分数Sarvam M (Reasoning, based on Mistral Small 3.1)Sarvam智能指数3→2.6-13%
- 分数Seed-OSS-36B-InstructByteDance Seed智能指数18.3→18.5+1.1%
- 分数Solar MiniUpstage智能指数6.2→6-3.2%
- 分数Solar Open 100B (Reasoning)Upstage智能指数15.1→15.2+0.7%
- 分数Solar Pro 2 (Non-reasoning)Upstage智能指数7.8→7.6-2.6%
- 分数Solar Pro 2 (Preview) (Non-reasoning)Upstage智能指数10→9.9-1.0%
- 分数Solar Pro 2 (Reasoning)Upstage智能指数9→8.8-2.2%
- 分数Solar Pro 3Upstage智能指数14.1→14.5+2.8%
- 分数Solar Pro 3Upstage智能体指数2.7→2.9+7.4%
- 分数SonarPerplexity智能指数9.5→9.4-1.1%
- 分数Sonar ProPerplexity智能指数9.3→9.1-2.2%
- 分数Sonar ReasoningPerplexity智能指数11.7→11.6-0.9%
- 分数Sonar Reasoning ProPerplexity智能指数17.8→18+1.1%
- 分数Step 3.5 FlashStepFun智能指数25.5→26+2.0%
- 分数Step 3.5 Flash 2603StepFun智能指数26→26.5+1.9%
- 分数Step 3.7 FlashStepFun智能指数30.3→30.9+2.0%
- 分数Step 3.7 FlashStepFun智能体指数21.5→21.7+0.9%
- 分数Step3 VL 10BStepFun智能指数9.5→9.3-2.1%
- 分数Tri-21B-ThinkTrillion Labs智能指数12.4→12.3-0.8%
- 分数Trinity Large ThinkingArcee AI智能指数18.2→18.4+1.1%
- 分数Trinity Large ThinkingArcee AI智能体指数3.7→指标撤下
08-05
2026年8月5日12 条变化- 下架Qwen3.8 MaxAlibaba
- 价格Celeris-1Celeris输入价$2.00→$0.200-90%
- 价格Celeris-1Celeris输出价$6.00→$0.700-88%
- 价格Gemma 4 31B (Non-reasoning)Google输入价$0.140→$0.150+7.1%
- 价格GLM-5.2 (max)Z AI输入价$1.40→$1.35-3.6%
- 价格GLM-5.2 (max)Z AI输出价$4.40→$4.29-2.5%
- 价格Grok 4.20 0309 v2 (Reasoning)SpaceXAI输入价$2.00→$1.25-38%
- 价格Grok 4.20 0309 v2 (Reasoning)SpaceXAI输出价$6.00→$2.50-58%
- 价格Hy3-preview (Non-reasoning)Tencent输出价$0.230→$0.210-8.7%
- 价格Hy3-preview (Reasoning)Tencent输出价$0.230→$0.210-8.7%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输入价$0.680→$0.600-12%
- 价格Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA输出价$2.67→$2.75+3.0%
08-04
2026年8月4日7 条变化- 新增G9v3-39A5BAI9Stars智能指数 30.9
- 新增Qwen3.8 MaxAlibaba智能指数 53.4
- 价格Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI输入价$2.00→$1.25-38%
- 价格Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI输出价$6.00→$2.50-58%
- 价格Inkling (xhigh)Thinking Machines输入价$1.87→$1.00-47%
- 价格Inkling (xhigh)Thinking Machines输出价$4.68→$4.05-13%
- 价格Llama 4 MaverickMeta输出价$0.850→$0.910+7.1%
08-01
2026年8月1日1 条变化- 新增Celeris-1Celeris智能指数 11.8
07-31
2026年7月31日49 条变化- 新增DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek智能指数 49.9
- 新增Inkling SmallThinking Machines智能指数 40.2
- 新增Kimi K3 (low)Kimi智能指数 46.6
- 价格GPT-5 (ChatGPT)OpenAI输入价$1.25→指标撤下
- 价格GPT-5 (ChatGPT)OpenAI输出价$10.00→指标撤下
- 价格GPT-5.6 Luna (high)OpenAI输入价$1.00→$0.200-80%
- 价格GPT-5.6 Luna (high)OpenAI输出价$6.00→$1.20-80%
- 价格GPT-5.6 Luna (low)OpenAI输入价$1.00→$0.200-80%
- 价格GPT-5.6 Luna (low)OpenAI输出价$6.00→$1.20-80%
- 价格GPT-5.6 Luna (max)OpenAI输入价$1.00→$0.200-80%
- 价格GPT-5.6 Luna (max)OpenAI输出价$6.00→$1.20-80%
- 价格GPT-5.6 Luna (medium)OpenAI输入价$1.00→$0.200-80%
- 价格GPT-5.6 Luna (medium)OpenAI输出价$6.00→$1.20-80%
- 价格GPT-5.6 Luna (Non-reasoning)OpenAI输入价$1.00→$0.200-80%
- 价格GPT-5.6 Luna (Non-reasoning)OpenAI输出价$6.00→$1.20-80%
- 价格GPT-5.6 Luna (xhigh)OpenAI输入价$1.00→$0.200-80%
- 价格GPT-5.6 Luna (xhigh)OpenAI输出价$6.00→$1.20-80%
- 价格GPT-5.6 Terra (high)OpenAI输入价$2.50→$2.00-20%
- 价格GPT-5.6 Terra (high)OpenAI输出价$15.00→$12.00-20%
- 价格GPT-5.6 Terra (low)OpenAI输入价$2.50→$2.00-20%
- 价格GPT-5.6 Terra (low)OpenAI输出价$15.00→$12.00-20%
- 价格GPT-5.6 Terra (max)OpenAI输入价$2.50→$2.00-20%
- 价格GPT-5.6 Terra (max)OpenAI输出价$15.00→$12.00-20%
- 价格GPT-5.6 Terra (medium)OpenAI输入价$2.50→$2.00-20%
- 价格GPT-5.6 Terra (medium)OpenAI输出价$15.00→$12.00-20%
- 价格GPT-5.6 Terra (Non-reasoning)OpenAI输入价$2.50→$2.00-20%
- 价格GPT-5.6 Terra (Non-reasoning)OpenAI输出价$15.00→$12.00-20%
- 价格GPT-5.6 Terra (xhigh)OpenAI输入价$2.50→$2.00-20%
- 价格GPT-5.6 Terra (xhigh)OpenAI输出价$15.00→$12.00-20%
- 价格gpt-oss-120b (low)OpenAI输出价$0.600→$0.590-1.7%
- 价格gpt-oss-20b (high)OpenAI输入价$0.060→$0.070+17%
- 价格gpt-oss-20b (low)OpenAI输出价$0.200→$0.230+15%
- 价格Hy3Tencent输出价$0.580→$0.560-3.4%
- 价格Llama 3.2 Instruct 11B (Vision)Meta输入价$0.360→$0.340-5.6%
- 价格Llama 3.2 Instruct 11B (Vision)Meta输出价$0.360→$0.340-5.6%
- 价格Llama 3.2 Instruct 90B (Vision)Meta输入价$2.04→指标撤下
- 价格Llama 3.2 Instruct 90B (Vision)Meta输出价$2.04→指标撤下
- 价格Llama 3.3 Instruct 70BMeta输入价$0.590→$0.600+1.7%
- 价格LongCat 2.0LongCat输入价首次评测→$0.750
- 价格LongCat 2.0LongCat输出价首次评测→$2.95
- 价格Nemotron 3 Super 120B A12B (Reasoning)NVIDIA输入价$0.250→$0.200-20%
- 价格Nemotron 3 Super 120B A12B (Reasoning)NVIDIA输出价$0.780→$0.800+2.6%
- 价格Solar Pro 3Upstage输入价首次评测→$0.150
- 价格Solar Pro 3Upstage输出价首次评测→$0.600
- 分数Gemma 4 E2B (Reasoning)Google智能指数9.4→9.7+3.2%
- 分数Gemma 4 E4B (Reasoning)Google智能指数11.9→12.2+2.5%
- 分数Ministral 3 3BMistral智能指数6.3→6.8+7.9%
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能指数5.3→5.5+3.8%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数7.1→7.6+7.0%
07-28
2026年7月28日14 条变化- 价格GLM-5.1 (Reasoning)Z AI输入价$1.39→$1.38-0.7%
- 价格Hy3-preview (Non-reasoning)Tencent输入价$0.120→$0.060-50%
- 价格Hy3-preview (Non-reasoning)Tencent输出价$0.430→$0.230-47%
- 价格Hy3-preview (Reasoning)Tencent输入价$0.120→$0.060-50%
- 价格Hy3-preview (Reasoning)Tencent输出价$0.430→$0.230-47%
- 价格Llama 3.3 Instruct 70BMeta输出价$0.710→$0.720+1.4%
- 价格Mistral MediumMistral输入价$2.75→$1.50-45%
- 价格Mistral MediumMistral输出价$8.10→$7.50-7.4%
- 价格Mistral Small (Feb '24)Mistral输入价$1.00→$0.150-85%
- 价格Mistral Small (Feb '24)Mistral输出价$3.00→$0.600-80%
- 分数Gemma 4 E2B (Reasoning)Google智能指数9.5→9.4-1.1%
- 分数Ministral 3 3BMistral智能指数6.5→6.3-3.1%
- 分数Qwen3.5 0.8B (Reasoning)Alibaba智能指数5.5→5.3-3.6%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数6.9→7.1+2.9%
07-25
2026年7月25日28 条变化- 新增Agnes 2.5 Pro AlphaSapiens AI智能指数 38.8
- 新增Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic智能指数 58.9
- 新增Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic智能指数 50.6
- 新增Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic智能指数 60.7
- 新增Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic智能指数 56.3
- 新增Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic智能指数 60.1
- 价格Gemma 3n E4B InstructGoogle输入价$0.020→$0.060+200%
- 价格Gemma 3n E4B InstructGoogle输出价$0.040→$0.120+200%
- 价格GLM-5.1 (Non-reasoning)Z AI输入价$1.40→$1.38-1.4%
- 价格GLM-5.1 (Reasoning)Z AI输入价$1.40→$1.39-0.7%
- 价格GLM-5.2 (Non-reasoning)Z AI输入价$1.40→$1.35-3.6%
- 价格GLM-5.2 (Non-reasoning)Z AI输出价$4.40→$4.25-3.4%
- 价格Grok 4SpaceXAI输入价$5.50→$3.00-45%
- 价格Grok 4SpaceXAI输出价$27.50→$15.00-45%
- 价格Llama 4 MaverickMeta输入价$0.310→$0.270-13%
- 价格MiMo-V2.5-Pro (Non-reasoning)Xiaomi输入价$0.660→$0.430-35%
- 价格MiMo-V2.5-Pro (Non-reasoning)Xiaomi输出价$1.72→$0.870-49%
- 价格Qwen3.5 9B (Reasoning)Alibaba输入价$0.100→$0.140+40%
- 价格Qwen3.5 9B (Reasoning)Alibaba输出价$0.150→$0.200+33%
- 分数EXAONE 4.5 33BLG AI Research智能指数23→20.2-12%
- 分数EXAONE 4.5 33BLG AI Research编程指数首次评测→23.6
- 分数EXAONE 4.5 33BLG AI Research智能体指数首次评测→9.7
- 分数Gemma 4 12B (Reasoning)Google智能指数22→21.8-0.9%
- 分数Gemma 4 12B (Reasoning)Google编程指数首次评测→31
- 分数Gemma 4 E2B (Reasoning)Google智能指数9.3→9.5+2.2%
- 分数Gemma 4 E4B (Reasoning)Google智能指数12.2→11.9-2.5%
- 分数Ministral 3 3BMistral智能指数6.8→6.5-4.4%
- 分数Qwen3.5 2B (Reasoning)Alibaba智能指数7.6→6.9-9.2%
07-24
2026年7月24日7 条变化- 新增G9v3-3BAI9Stars智能指数 16.1
- 分数K-EXAONE (Reasoning)LG AI Research智能指数24.7→22.1-11%
- 分数K-EXAONE (Reasoning)LG AI Research编程指数首次评测→32.1
- 分数K-EXAONE (Reasoning)LG AI Research智能体指数首次评测→8
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能指数21.3→17.6-17%
- 分数Nemotron Cascade 2 30B A3BNVIDIA编程指数首次评测→25.3
- 分数Nemotron Cascade 2 30B A3BNVIDIA智能体指数首次评测→3.7