Science Question Answering on GPQA Diamond (avg@16, best@16)
66.48Average@16THINC-4B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| THINC-4B2026.05 | 66.48 | 91.41 | |
| Qwen3-4B-Thinking2026.05 | 66.32 | 83.84 | |
| ASTER-4B2026.05 | 63.42 | 90.4 |
| Method | Links | ||
|---|---|---|---|
| THINC-4B2026.05 | 66.48 | 91.41 | |
| Qwen3-4B-Thinking2026.05 | 66.32 | 83.84 | |
| ASTER-4B2026.05 | 63.42 | 90.4 |