DeepSeek V4 Flash 0731 High
| Released | 2026-07-31 |
|---|---|
| Price per 1M | $0.15 in · $0.30 out |
| Overall score | 57.4 · rank 160 |
| Context window | 1,000,000 tokens |
| Provider | DeepSeek |
| Weights | Open weights (MIT) |
| Configuration | High |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 57.4 | 160 | 18 |
| Overall: Coding | 57.0 | 167 | 15 |
| Overall: Reasoning | 58.1 | 149 | 7 |
| Overall: Agentic work | 58.4 | 151 | 14 |
| Overall: Frontend | 56.8 | 137 | 6 |
| Coding | 55.5 | 170 | 5 |
| Agents | 59.3 | 90 | 5 |
| Reasoning | 59.6 | 138 | 4 |
| Math | 58.7 | 66 | 1 |
| Knowledge | 53.9 | 88 | 2 |
| Frontend | 57.5 | 69 | 1 |
| Value vs price | +11.9 | 33 |
Evidence
Coding 55.5 ±2.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierCode | 18.8% | Epoch AI | 2026-11-11 |
| IOI | 32.7% | Vals AI | 2026-10-29 |
| SWE-bench Verified (Vals run) | 88.8% | Vals AI | 2026-09-01 |
| LiveCodeBench (Vals run) | 87.3% | Vals AI | 2026-09-01 |
| ProgramBench | 0.0% | Vals AI | 2026-11-09 |
Agents 59.3 ±1.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4.0 (Vals run) | 18.7% | Vals AI | 2026-11-09 |
| Terminal-Bench-Science | 4.3% | Vals AI | 2026-11-09 |
| Tax Agent Bench | 28.1% | Vals AI | 2026-11-09 |
| SkillsBench | 50.7% | Vals AI | 2026-10-29 |
| CyberBench v1.1 | 72.1% | Vals AI | 2026-11-05 |
Reasoning 59.6 ±2.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond (Vals run) | 89.9% | Vals AI | 2026-09-01 |
| ARC-AGI-2 | 56.0% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 87.0% | ARC Prize | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 56.0% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 87.0% | Epoch AI | 2026-11-11 |
| BioMysteryBench | 64.4% | Vals AI | 2026-11-09 |
Math 58.7 ±2.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ProofBench v1.1 | 56.0% | Vals AI | 2026-11-09 |
Knowledge 53.9 ±3.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 86.2% | Vals AI | 2026-09-01 |
| Legal Research Bench | 30.3% | Vals AI | 2026-11-09 |
Frontend 57.5 ±0.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 74.7% | Vals AI | 2026-11-09 |
Names matched to this model
deepseek-v4-flash-0731-high