DeepSeek-V3
| Released | 2024-12-26 |
|---|---|
| Price per 1M | $0.29 in · $1.16 out |
| Overall score | 40.1 · rank 193 |
| Context window | 131,072 tokens |
| Provider | DeepSeek |
| Weights | Open weights (DeepSeek Model License) |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 40.1 | 193 | 11 |
| Overall: Coding | 41.1 | 183 | 6 |
| Overall: Reasoning | 38.6 | 202 | 7 |
| Overall: Agentic work | 40.3 | 185 | 6 |
| Overall: Frontend | 41.8 | 167 | 2 |
| Coding | 41.8 | 173 | 2 |
| Agents | 39.8 | 157 | 1 |
| Reasoning | 39.7 | 198 | 3 |
| Math | 39.1 | 166 | 3 |
| Knowledge | 35.6 | 152 | 1 |
| General | 42.6 | 149 | 1 |
| Value vs price | −5.6 | 160 |
Evidence
Coding 41.8 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 48.4% | Epoch AI | 2026-11-11 |
| SciCode | 35.8% | Epoch AI | 2026-11-11 |
Agents 39.8 ±3.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| METR task suite | 47.4% | Epoch AI | 2026-11-11 |
Reasoning 39.7 ±3.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CritPt | 0.0% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 56.5% | Epoch AI | 2025-01-27 |
| GPQA Diamond (Vals run) | 54.5% | Vals AI | 2026-09-01 |
| SimpleBench | 27.2% | SimpleBench | 2026-11-11 |
| SimpleBench | 18.9% | Epoch AI | 2026-11-11 |
Math 39.1 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 (2025 set) | 1.7% | Epoch AI | 2025-03-07 |
| MATH Level 5 | 64.9% | Epoch AI | 2025-01-27 |
| OTIS Mock AIME 2024-2025 | 15.8% | Epoch AI | 2025-02-25 |
Knowledge 35.6 ±4.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 73.8% | Vals AI | 2026-09-01 |
General 42.6 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,358 | LMArena | 2026-11-09 |
Names matched to this model
deepseek-v3, deepseek-v3-base