Model card
DeepSeek-V2.5.
DeepSeekopen-sourceLLM
DeepSeek's V2.5 model.
§ 01 · Benchmarks
Every benchmark DeepSeek-V2.5 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | SWE-Bench Verified | Computer Code · Code Generation | resolve-rate | 37.0% | #39 | — | source ↗ |
| 02 | SWE-bench Verified | Agentic AI · SWE-bench | resolve-rate | 16.8% | #81 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where DeepSeek-V2.5 actually performs.
§ 04 · Related models
Other DeepSeek models scored on Codesota.
§ 05 · Sources & freshness
Where these numbers come from.
deepseek-blog
1
result
editorial
1
result
1 of 2 rows marked verified.