DeepSeekAuto-detectedDeepSeek LLM 67B
Best configuration ranks #587 of 610 on the BenchLeader Index at 36.0 ±3.6. Last measured 2 Sept 2026. Released 29 Nov 2023.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- 4k
- Overall index36
- Reasoning24
- Coding32
- Maths25
- Human preference32
- Composite36
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source |
|---|---|---|
| GPQA Diamond | 24.6%#263 | Epoch AI Benchmarking Hub |
| LMArena Hard Prompts | 1175#302 | LMArena |
Coding
| Benchmark | default | Source |
|---|---|---|
| LMArena Coding | 1218#294 | LMArena |
Maths
| Benchmark | default | Source |
|---|---|---|
| OTIS Mock AIME | 0.8%#249 | Epoch AI Benchmarking Hub |
| MATH Level 5 | 6.4%#85 | Epoch AI Benchmarking Hub |
Human preference
| Benchmark | default | Source |
|---|---|---|
| LMArena Text | 1184#291 | LMArena |
Composite
| Benchmark | default | Source |
|---|---|---|
| Epoch Capabilities Indexnot in index | 110.5#189 | Epoch AI Benchmarking Hub |
| AA Intelligence Index | 5.3#496 | Artificial Analysis |
See also
Data as of 9 Sept 2026. Compare with another model.