DeepSeek-R1
| Released | 2025-01-20 |
|---|---|
| Price per 1M | $1.35 in · $5.40 out |
| Overall score | 46.0 · rank 150 |
| Context window | 128,000 tokens |
| Provider | DeepSeek |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 46.0 | 150 | 17 |
| Overall: Coding | 44.9 | 155 | 11 |
| Overall: Reasoning | 46.5 | 145 | 9 |
| Overall: Agentic work | 44.2 | 165 | 11 |
| Overall: Frontend | 45.3 | 132 | 3 |
| Coding | 45.3 | 150 | 3 |
| Agents | 43.3 | 143 | 2 |
| Reasoning | 45.9 | 146 | 6 |
| Math | 46.6 | 139 | 2 |
| Knowledge | 48.0 | 103 | 1 |
| General | 50.4 | 101 | 3 |
| Value vs price | −4.4 | 156 |
Evidence
Coding 45.3 ±2.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SciCode | 38.3% | Epoch AI | 2026-11-11 |
| Aider Polyglot | 56.9% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 70.2% | Vals AI | 2026-09-01 |
Agents 43.3 ±3.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| METR task suite | 51.9% | Epoch AI | 2026-11-11 |
| BALROG | 34.9% | Epoch AI | 2026-11-11 |
Reasoning 45.9 ±1.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 1.3% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 1.3% | ARC Prize | 2026-11-11 |
| GPQA Diamond | 71.7% | Epoch AI | 2025-05-26 |
| CritPt | 1.1% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 21.2% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 15.8% | Epoch AI | 2026-11-11 |
| Kagi Reasoning Benchmark | 69.4% | Kagi | 2026-11-11 |
| SimpleBench | 40.8% | SimpleBench | 2026-11-11 |
| SimpleBench | 30.9% | Epoch AI | 2026-11-11 |
Math 46.6 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 93.1% | Epoch AI | 2025-01-31 |
| OTIS Mock AIME 2024-2025 | 53.3% | Epoch AI | 2025-02-26 |
Knowledge 48.0 ±4.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 83.2% | Vals AI | 2026-09-01 |
General 50.4 ±3.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,398 | LMArena | 2026-11-09 |
| EQ-Bench Creative Writing v3 | 1,500 | EQ-Bench | 2026-11-11 |
| EQ-Bench Longform Creative Writing | 59.5% | EQ-Bench | 2026-11-11 |
Names matched to this model
deepseek-r1