LLM Tutoring Evaluation on TutorBench
64.1Text-Only ScoreEDUQWEN 30B-VL-SFT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| EDUQWEN 30B-VL-SFTJudge Model=CLAUDE-4.5-SONNET2026.04 | 64.1 | 59.4 | 61.64 | — | |
| QWEN3-30B-VL (BASE MODEL)Judge Model=CLAUDE-4.5-SONNET2026.04 | 60.21 | 51.53 | 55.39 | — | |
| GEMINI-3PROJudge Model=CLAUDE-4.5-SONNET2026.04 | 58.41 | 58.67 | 58.52 | — | |
| GEMINI-2.5-PRO-PREVIEW-06-05 (RANK #1)Judge Model=CLAUDE-4-SONNET2026.04 | — | — | — | 55.65 | |
| GEMINI-3-PRO-PREVIEW (RANK #3)Judge Model=CLAUDE-4-SONNET2026.04 | — | — | — | 53.67 |