DeepSeek V2.5 — Benchmarks
Benchmark scores for DeepSeek V2.5 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Coding
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| Aider Polyglot | 17.8% | #13 / 18 | #60 / 69 |
Knowledge
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| MMLU-Pro | 65.8% | #41 / 119 | #117 / 259 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.