Closed-ended Task T-7 on DanmakuTPP-QA
23AccuracyQwen3-30B-A3B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-30B-A3BModel Category=LLMs2026.02 | 23 | |
| Qwen3-8BModel Category=LLMs2026.02 | 19.33 | |
| Llama-3.3-70B-InstructModel Category=LLMs2026.02 | 17 | |
| Qwen2.5-32B-InstructModel Category=LLMs2026.02 | 16.67 | |
| Qwen2.5-72B-InstructModel Category=LLMs2026.02 | 16 | |
| Qwen2.5-VL-72BModel Category=MLLMs2026.02 | 15.98 | |
| Gemma3-27BModel Category=MLLMs2026.02 | 15.67 | |
| MM-TPPModel Category=Finetuned2026.02 | 15.67 | |
| Qwen2.5-VL-3BModel Category=Finetuned2026.02 | 15.33 | |
| DeepSeek-V3Model Category=LLMs2026.02 | 13.67 | |
| Qwen2.5-VL-32BModel Category=MLLMs2026.02 | 12.67 | |
| Qwen2.5-7B-InstructModel Category=LLMs2026.02 | 10.67 | |
| Qwen3-235B-A22BModel Category=LLMs2026.02 | 10.33 | |
| Qwen2.5-VL-7BModel Category=MLLMs2026.02 | 8.33 |