Actions on LiC Actions
99Concat ScoreQwen3-8B
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen3-8Bsource=Grattafiori et al.2025.10 | 99 | 48.5 | 49 | 34.7 | |
| GPT-4.12025.10 | 98.5 | 62.9 | 63.9 | 30.1 | |
| Llama4-Scout2025.10 | 98.3 | 69.9 | 71.1 | — | |
| Gemini-2.5-Pro2025.10 | 98.1 | 36.3 | 37 | 32.9 | |
| Qwen3-8BOptimization=RLAAR2025.10 | 97.4 | 52.3 | 53.7 | 69.7 | |
| GPT-4o2025.10 | 97.1 | 65 | 66.9 | 26.8 | |
| Llama3.3-70B2025.10 | 97 | 71 | 73.2 | — | |
| DeepSeek-R12025.10 | 97 | 47.5 | 49 | — | |
| Claude3.7-Sonnet2025.10 | 96 | 33.3 | 34.7 | — | |
| Qwen2.5-7B2025.10 | 95.8 | 35 | 36.5 | 31.9 | |
| Qwen2.5-7BOptimization=RLAAR2025.10 | 92.8 | 49.3 | 53 | 70.3 | |
| GPT4o-mini2025.10 | 92.2 | 52.4 | 56.8 | — | |
| Qwen2.5-7BFinetuning=SFT, Distillation=Gemini-2.5-Pro2025.10 | 91.7 | 31.6 | 34.5 | 52.7 | |
| GPT-o32025.10 | 91.5 | 60.2 | 65.8 | — | |
| Qwen3-8BFinetuning=SFT, Distillation=Gemini-2.5-Pro2025.10 | 90.2 | 41.6 | 46.1 | 47.8 | |
| Gemini-2.5-Flash2025.10 | 89.2 | 42.6 | 47.8 | 25.4 | |
| Qwen3-1.7BOptimization=RLAAR2025.10 | 84.8 | 41.3 | 48.7 | 65.2 | |
| Qwen3-1.7Bsource=Grattafiori et al.2025.10 | 83.1 | 40.2 | 48.4 | 24.9 | |
| Claude3-Haiku2025.10 | 80.2 | 55.9 | 69.7 | — | |
| Phi-42025.10 | 76 | 34.1 | 44.9 | — | |
| OLMo22025.10 | 49.8 | 13.8 | 27.7 | — |