Benchmark
AlpacaEval 2.0
AlpacaEval 2.0 is a length-controlled automatic evaluator for instruction-following language models that uses GPT-4 Turbo to assess model responses…
- Modality
- text
- Categories
- reasoning, general, creativity, writing
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 4
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek-V2.5 | DeepSeek | 0.505 | 2024-05-08 | evidence |
| Granite 3.3 8B Base | IBM | 0.6268 | 2025-04-16 | evidence |
| Granite 3.3 8B Instruct | IBM | 0.6268 | 2025-04-16 | evidence |
| IBM Granite 4.0 Tiny Preview | IBM | 0.3516 | 2025-05-02 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.