Claude 3.7 Sonnet scores 17.0 on the Respan Index (#112), ahead of Gemma 4 31B at 16.8 (#113). Gemma 4 31B leads on 4 of the 4 benchmarks both report. Gemma 4 31B has the larger context window (256K tokens).
| Benchmark | Claude 3.7 Sonnet | Gemma 4 31B |
|---|---|---|
| GPQA Diamond | 79.7% | 84.3% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 57.8% | 73.3% |
| LMArena Elo | 1388 | 1442 |
| LMArena Vision (LMArena) | 1193 | 1262 |
Higher is better. Each score links to where it was published.
* Estimated: no published scores in that category. How the Respan Index works