DeepSeek V4 Flash 0731 ranks first on Long-Horizon-Terminal-Bench (LHTB).
DeepSeek Jul 2026 Open weights Mid-size
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BullshitBench | 40% | #37 of 107 | Official leaderboard | |
| BullshitBench V2 | 39% | #59 of 122 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 61.2 | #60 of 119 | Official leaderboard | |
| Creative Writing v3 | 67.5 | #62 of 120 | Official leaderboard | |
| SimpleQA Verified | 33.6% | #60 of 84 | Epoch AI | |
| Judgemark v4 | 30.8 | #35 of 40 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 91% | #32 of 219 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 94.4% | #46 of 202 | Epoch AI | |
| DTBench | 84.9% | #41 of 178 | Official leaderboard | |
| CritPt | 16.6% | #42 of 160 | Epoch AI | |
| SimpleBench | 61.1% | #22 of 82 | Epoch AI | |
| ARC-AGI-2 | 61.4% | #30 of 95 | Official leaderboard | |
| ProofBench | 39% | #19 of 59 | Official leaderboard | |
| ARC-AGI | 89% | #35 of 92 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 57.5% | #45 of 89 | Epoch AI | |
| ProofBench v1.1 | 56% | #27 of 51 | Official leaderboard | |
| FrontierMath Tier 4 v2 | 24.4% | #46 of 71 | Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WebDev Arena | 1,580 | #25 of 105 | Epoch AI | |
| WeirdML | 63% | #27 of 112 | Official leaderboard | |
| ALE-Bench | 1,306 | #33 of 121 | Epoch AI | |
| Vibe Code Bench | 74.7% | #39 of 115 | Official leaderboard | |
| NL2Repo-Bench | 54.2% | #10 of 28 | Model card | |
| Code Migration | 38.6% | #33 of 78 | Official leaderboard | |
| DSBench (DeepSeek internal) | 68.7% | #2 of 3 | Model card | |
| DeepSWE | 54.4% | #41 of 51 | Model card | |
| FrontierCode | 18.8% | #39 of 46 | Epoch AI | |
| IOI v2 (Vals) | 32.7% | #40 of 45 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Long-Horizon-Terminal-Bench (LHTB) | 60.2% | #1 of 25 | Official leaderboard | |
| Toolathlon | 70.3% | #9 of 37 | Model card | |
| Mystery Game Puzzles | 34% | #25 of 80 | Epoch AI | |
| Toolathlon-Verified | 70.7% | #9 of 25 | Official leaderboard | |
| GAIA | 66.5% | #21 of 57 | Official leaderboard | |
| Terminal-Bench 2.1 | 82.7% | #22 of 53 | Model card | |
| Chess Puzzles | 32% | #48 of 83 | Epoch AI | |
| AutomationBench | 25.1% | #33 of 41 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Harvey LAB (Legal Agent Benchmark) | 8.33% | #26 of 80 | Official leaderboard | |
| Tax Agent Bench | 28.1% | #34 of 79 | Official leaderboard | |
| CorpFin v2 | 61.9% | #53 of 119 | Official leaderboard | |
| PolicyBench | 82.9% | #26 of 50 | Official leaderboard | |
| MedCode | 41.4% | #51 of 96 | Official leaderboard | |
| MedScribe | 80.4% | #54 of 98 | Official leaderboard | |
| Excel Modeling Benchmark | 57% | #43 of 76 | Official leaderboard | |
| Legal Research Bench | 30.3% | #51 of 88 | Official leaderboard | |
| Vals Finance Agent v2 | 49.5% | #47 of 80 | Official leaderboard | |
| TaxEval v2 | 70.7% | #82 of 129 | Official leaderboard | |
| LegalBench | 77.7% | #100 of 138 | Official leaderboard | |
| Agents' Last Exam | 25.2% | #21 of 25 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Code2Video Benchmark | 1,472 | #13 of 19 | Official leaderboard | |
| VoxelBench | 1,401 | #43 of 56 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CyberBench (Vals) | 77.5% | #6 of 26 | Official leaderboard | |
| CyberBench v1.1 (Vals) | 72.1% | #16 of 43 | Official leaderboard | |
| CyberGym | 76.7% | #20 of 33 | Model card | |
| CyberGym-E2E | 28.4% | #5 of 5 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMCA | 41.7% | #47 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 54.4 | #53 of 149 | Official leaderboard | |
| ACCoRD | 61.7 | #84 of 150 | Official leaderboard |
DeepSeek's models in color, against every other model on the Epoch Capabilities Index, by release date.
The benchmarks the lab chose to headline, with the scores it gave.