Long-context graph reasoning on GraphWalks (F1)
82.3F1 ScoreGemma 4 31B
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemma 4 31BModel Size/Architecture=31B, Context length=<128k, Thinking mode=without thinking2026.07 | 82.3 | |
| Gemma 4 26B-A4BModel Size/Architecture=26B-A4B, Context length=<128k, Thinking mode=without thinking2026.07 | 72.6 | |
| Gemma 4 12BModel Size/Architecture=12B, Context length=<128k, Thinking mode=without thinking2026.07 | 71 | |
| Gemma 4 E4BModel Size/Architecture=E4B, Context length=<128k, Thinking mode=without thinking2026.07 | 50.9 | |
| Gemma 3 27BModel Size/Architecture=27B, Context length=<128k, Thinking mode=without thinking2026.07 | 32.8 | |
| Gemma 4 E2BModel Size/Architecture=E2B, Context length=<128k, Thinking mode=without thinking2026.07 | 4.1 |