Criterion-Conditional In-Context Learning on CC-Bench Industrial Inspection 1.0 (test)
29.74CS ScoreGPT-4o
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| GPT-4oModel Category=Proprietary API Models, Support Set=3-shot2026.06 | 29.74 | 82.46 | 58 | |
| Gemini 2.5 ProModel Category=Proprietary API Models, Support Set=3-shot2026.06 | 20.04 | 88.99 | 57 | |
| Qwen2.5-VL-32B-InstructModel Category=Open-Source Models, Support Set=3-shot2026.06 | 17.67 | 58.58 | 39.6 | |
| InternVL3.5-8BModel Category=Open-Source Models, Support Set=3-shot2026.06 | 16.81 | 70.15 | 45.4 | |
| LLaVA-OneVision-1.5-8B-InstructModel Category=Open-Source Models, Support Set=3-shot2026.06 | 15.95 | 38.81 | 28.2 | |
| Qwen2.5-VL-7B-InstructModel Category=Open-Source Models, Support Set=3-shot2026.06 | 13.79 | 39.74 | 27.7 | |
| Qwen3-VL-8B-InstructModel Category=Open-Source Models, Support Set=3-shot2026.06 | 11.21 | 83.02 | 49.7 |