← All benchmarks
DeepSeek V4 Pro (Reasoning, High Effort)
DeepSeek · Released 2026-04-24 · Independent benchmark scores from Artificial Analysis
Intelligence Index
43.7
Output speed
61.2
tokens/sec
Blended price
$0.54
per 1M tokens
Time to first token
1.37s
Model metadata
| Model ID | 5da3c0e2-65d2-4bff-a410-cb2132ddafb6 |
|---|---|
| Slug | deepseek-v4-pro-high |
| Release date | 2026-04-24 |
| Provider | DeepSeek |
| Provider slug | deepseek |
| Output speed | 61.2 tok/s |
| Time to first token | 1.37s |
| Time to first answer | 33.92s |
| Blended price ($/1M) | $0.54 |
| Input price ($/1M) | $0.43 |
| Output price ($/1M) | $0.87 |
All evaluation scores (17)
| aime | — |
|---|---|
| aime 25 | — |
| artificial analysis coding index | 58.70 |
| artificial analysis intelligence index | 43.70 |
| artificial analysis math index | — |
| gpqa | 90.5% |
| hle | 35.2% |
| ifbench | 71.3% |
| lcr | 67.0% |
| livecodebench | — |
| math 500 | — |
| mmlu pro | — |
| scicode | 46.4% |
| tau banking | 26.2% |
| tau2 | 94.2% |
| terminalbench hard | 41.7% |
| terminalbench v2 1 | 64.8% |
Source: artificialanalysis.ai/models/deepseek-v4-pro-high
Benchmark data from Artificial Analysis.