Leaderboard / Model
DeepSeek V4 Flash 0731
DeepSeek · Open weights · Released Jul 31, 2026 · Price not listed
n/aOverall · not enough data
Score by job
Coding
Not enough data · 1 of 5 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| FrontierCode | 18.8% | deepseek-v4-flash-0731_unknown | -2.16 |
Agents
Not enough data · 0 of 4 benchmarks
Reasoning
42 · rank 26 · 3 of 4 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| GPQA Diamond | 91.0% | deepseek-v4-flash-0731_max | -0.21 |
| ARC-AGI-2 | 61.4% | deepseek-v4-flash-0731_max | -0.12 |
| SimpleBench | 61.1% | deepseek-v4-flash-0731_unknown | -0.43 |
Math
34 · rank 25 · 4 of 5 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| FrontierMath Tiers 1–3 | 57.5% | deepseek-v4-flash-0731_max | -0.72 |
| FrontierMath Tier 4 | 24.4% | deepseek-v4-flash-0731_max | -0.69 |
| OTIS Mock AIME | 94.4% | deepseek-v4-flash-0731_max | -0.28 |
| ProofBench | 56.0% | deepseek-v4-flash-0731_unknown | 0.33 |
Writing
Not enough data · 0 of 2 benchmarks
Design
Not enough data · 0 of 1 benchmarks