DeepSeek R1 Distill Llama 70B — Benchmarks
Benchmark scores for DeepSeek R1 Distill Llama 70B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| AIME 2024/2025 | 51.4% | #13 / 34 | #87 / 155 |
| MATH Level 5 | 89.9% | #3 / 32 | #23 / 108 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| GPQA Diamond | 55.7% | #18 / 46 | #112 / 182 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.