DeepSeek
DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
Quality Score63.3
Value Score47.9
Reliability50.4
Cache Discount97%
Model specification
- Reasoning
- reasoning
- Input modalities
- text
- Output modalities
- text
- Context window
- 1M tokens
- Weights
- open weights
Token prices USD per 1M tokens
- Input
- $1.32
- Output
- $3.96
- Cached Input
- $0.044
- Cached Output
- not available
- Cost per task
- $0.252
Capability
- Intelligence
- 53.2
- Coding
- 68.8
- Agentic
- 49.6
- Omniscience
- 0.8
- Correct
- 49.1%
- Blended price
- $0.691
Answer outcomes
Correct 49.1%Incorrect 48.3%Abstained 2.6%
Artificial Analysis benchmarks
- GDPval-AA v2
- 55%
- τ³-Banking
- 40%
- Terminal-Bench v2.1
- 79%
- SciCode
- 49%
- Humanity’s Last Exam
- 41%
- GPQA Diamond
- 93%
- CritPt
- 18%
- AA-Omniscience
- 50%
- AA-LCR
- 75%
Similar models
- DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeekQuality Score 55.1Value Score 50.5
- DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeekQuality Score 52.0Value Score 47.6
- DeepSeek V4 Pro (Reasoning, High Effort)DeepSeekQuality Score 50.6Value Score 46.0
- DeepSeek V4 Pro (Non-reasoning)DeepSeekQuality Score 31.8Value Score 33.5
- DeepSeek V4 Flash (Non-reasoning)DeepSeekQuality Score 25.5Value Score 31.6
- Claude Sonnet 5 (Adaptive Reasoning, Max Effort)AnthropicQuality Score 64.1Value Score 43.6