DeepSeekDeepSeek-R1-Distill-Llama-70B
Best configuration ranks #464 of 610 on the BenchLeader Index at 42.2 ±6.6. Last measured 20 Jan 2025stale: no new result in six months. Released 20 Jan 2025.
- Blended price
- $0.815/M
- $0.800 in · $0.861 out
- Output speed
- 21 tok/s
- First answer
- 96 s
- first token 1.05 s
- Context
- 33k
- Overall index42
- Reasoning44
- Agents & tools35
- Maths56
- Knowledge42
- Instruction following32
- Long context30
- Composite39
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 | 55.7%#188 | Epoch AI Benchmarking Hub |
| GPQA Diamond (AA)not in index | 40.2%#453 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 5.1%#368 | Artificial Analysis |
| Kagi LLM Benchmark | 52.3%#73 | Kagi LLM Benchmark |
Agents & tools
| Benchmark | default | Source |
|---|---|---|
| Terminal-Bench Hard | 1.5%#321 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 21.9%#303 | Artificial Analysis |
Maths
| Benchmark | default | Source |
|---|---|---|
| OTIS Mock AIME | 51.4%#164 | Epoch AI Benchmarking Hub |
| MATH Level 5 | 89.9%#20 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| AA-Omniscience | -46.4#291 | Artificial Analysis |
Instruction following
| Benchmark | default | Source |
|---|---|---|
| IFBench | 27.6%#360 | Artificial Analysis |
Long context
| Benchmark | default | Source |
|---|---|---|
| AA-LCR | 10.0%#395 | Artificial Analysis |
Composite
| Benchmark | default | Source |
|---|---|---|
| AA Intelligence Index | 7.9#374 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| NovitaAI | 18 tok/s | 1.05 s | $0.800 | $0.800 | 8k | bf16 |
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 | $0.0006 | – | 1.8 min |
| Summarise a 30-page report | 12,000 / 600 | $0.010 | – | 2.1 min |
| Code edit | 6,000 / 1,500 | $0.0061 | – | 2.8 min |
| Agentic coding session | 60,000 / 4,000 | $0.051 | – | 4.8 min |
| Structured extraction | 2,000 / 200 | $0.0018 | – | 1.8 min |
See also
Data as of 9 Sept 2026. Compare with another model.