DeepSeekDeepSeek V4 Pro 0813 (Reasoning, Max Effort)
- Released
- Aug 13, 2026
- Tracked since
- 2026-08-13
Intelligence index36Ranks 64 of 664 models with this metric
Coding index68.8Ranks 56 of 259 models with this metric
Agentic index41.3Ranks 39 of 156 models with this metric
Capability profile
5 axes vs. site averageDeepSeek V4 Pro 0813 (Reasoning, Max Effort)
Site average
Pricing
per 1M tokens- Input
- $1.32
- Output
- $3.96
- Cache read
- $0.040
Performance
- Output throughput
- 89 tokens/sec
- Time to first token
- 1.69s
- First answer token
- 24.14s
- End-to-end response
- 29.76s
History
8Intelligence index
53→36
Output
$0.870→$3.96
Output throughput
83→89
Cheaper at this level
Models with a comparable intelligence index and a lower output price- Qwen3.8-Flash-Next
AlibabaIntelligence 39.8$0.470−88% - GLM 5.3 Flash
Z AIIntelligence 41.8$0.500−87% - GPT-6 Luna (max)
OpenAIIntelligence 37.3$0.500−87% - MiMo-V2.6-Pro
XiaomiIntelligence 46.3$0.870−78% - DeepSeek V4.1 Flash (Reasoning, Max Effort)DeepSeekIntelligence 39.5$1.20−70%