deepseek-ai/DeepSeek-V2.5
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| HumanEval | coding | 84.5 | HumanEval python | medium | | HumanEval python | 84.5 | 87.2 | 89 | |
| HumanEval | coding | 73.8 | HumanEval Multi | medium | | HumanEval Multi | 73.8 | 74.8 | 73.8 | |
| Chatbot Arena | chat | 68.3 | ArenaHard | medium | | ArenaHard | 68.3 | 66.3 | 76.2 | |
| AlpacaEval | chat | 46.6 | AlpacaEval 2.0 | medium | | AlpacaEval 2.0 | 46.6 | 44.5 | 50.5 | |
| LiveCodeBench | coding | 36.6 | LiveCodeBench(01-09) | medium | | LiveCodeBench(01-09) | 36.6 | 39.7 | 41.8 | |
| MT-Bench | chat | 8.85 | MT-Bench | medium | | MT-Bench | 8.85 | 8.91 | 9.02 | |
| AlpacaEval | chat | 2.0 | medium | AlpacaEval 2.0 |
none scraped.