DeepSeek V3 0324
| Released | 2025-03-24 |
|---|---|
| Price per 1M | $0.29 in · $1.14 out |
| Overall score | 43.1 · rank 175 |
| Context window | 163,840 tokens |
| Provider | DeepSeek |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 43.1 | 175 | 11 |
| Overall: Coding | 43.5 | 168 | 7 |
| Overall: Reasoning | 41.7 | 183 | 6 |
| Overall: Agentic work | 42.1 | 180 | 7 |
| Overall: Frontend | 44.7 | 140 | 3 |
| Coding | 44.7 | 158 | 3 |
| Agents | 41.0 | 153 | 1 |
| Reasoning | 41.7 | 177 | 3 |
| Math | 41.2 | 160 | 2 |
| Knowledge | 42.6 | 130 | 1 |
| General | 47.8 | 130 | 1 |
| Value vs price | −2.6 | 148 |
Evidence
Coding 44.7 ±2.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SciCode | 39.0% | Epoch AI | 2026-11-11 |
| Aider Polyglot | 55.1% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 65.5% | Vals AI | 2026-09-01 |
Agents 41.0 ±3.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| METR task suite | 49.6% | Epoch AI | 2026-11-11 |
Reasoning 41.7 ±3.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CritPt | 0.0% | Epoch AI | 2026-11-11 |
| SimpleBench | 27.2% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 67.6% | Epoch AI | 2025-04-01 |
| GPQA Diamond (Vals run) | 61.6% | Vals AI | 2026-09-01 |
Math 41.2 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 75.6% | Epoch AI | 2025-04-01 |
| OTIS Mock AIME 2024-2025 | 37.8% | Epoch AI | 2025-04-01 |
Knowledge 42.6 ±4.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 79.5% | Vals AI | 2026-09-01 |
General 47.8 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,395 | LMArena | 2026-11-09 |
Names matched to this model
deepseek-chat-v3-0324, deepseek-v3-0324