A compact, nimble model that punches above its weight for its size. Gemma 2 2B handles everyday language tasks — summarization, Q&A, light reasoning — with surprising coherence, though it naturally hits walls on complex multi-step problems that larger models handle more gracefully. It runs efficiently on modest hardware, making it practical for edge deployments or resource-constrained environments.
| Benchmark | Score | Type | Recorded |
|---|---|---|---|
| MuSR | 7.1 | accuracy | 24d ago |
| IFEval | 56.7 | accuracy | 24d ago |
| BBH | 18.0 | accuracy | 24d ago |
| MMLU-Pro | 17.2 | accuracy | 24d ago |
| GPQA Diamond | 3.2 | accuracy | 24d ago |
| MATH | 0.1 | accuracy | 24d ago |