DeepSeekDeepSeek-R1-Distill-Qwen-32B
Best configuration ranks #454 of 610 on the BenchLeader Index at 42.8 ±7.7. Last measured 20 Jan 2025stale: no new result in six months. Released 20 Jan 2025.
- Blended price
- $0.440/M
- $0.300 in · $0.861 out
- Output speed
- –
- First answer
- –
- Context
- 64k
- Overall index43
- Reasoning48
- Maths48
- Instruction following28
- Long context29
- Composite40
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 | 64.1%#170 | Epoch AI Benchmarking Hub |
| GPQA Diamond (AA)not in index | 61.5%#347 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 4.6%#412 | Artificial Analysis |
Maths
| Benchmark | default | Source |
|---|---|---|
| OTIS Mock AIME | 55.6%#157 | Epoch AI Benchmarking Hub |
Instruction following
| Benchmark | default | Source |
|---|---|---|
| IFBench | 22.9%#381 | Artificial Analysis |
Long context
| Benchmark | default | Source |
|---|---|---|
| AA-LCR | 8.7%#399 | Artificial Analysis |
Composite
| Benchmark | default | Source |
|---|---|---|
| Epoch Capabilities Indexnot in index | 137.4#109 | Epoch AI Benchmarking Hub |
| AA Intelligence Index | 8.4#363 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0004 | – | – |
| Summarise a 30-page report | 12,000 / 600 | $0.0041 | – | – |
| Code edit | 6,000 / 1,500 | $0.0031 | – | – |
| Agentic coding session | 60,000 / 4,000 | $0.021 | – | – |
| Structured extraction | 2,000 / 200 | $0.0008 | – | – |
See also
Data as of 9 Sept 2026. Compare with another model.