General Reasoning on RACE
86.51AccuracyQwen2.5-7B
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2.5-7B2026.02 | 86.51 | |
| BioBridge2026.02 | 84 | |
| Sumi-7BParadigm=Uniform Diffusion, Training Tokens=1.5T, Training Data=Fully Released, Evaluation protocol=Evaluated under our protocol, Shots=02026.06 | 41.4 | |
| Llama 2-7BParadigm=AR, Training Tokens=2T, Training Data=Not Released, Evaluation protocol=Evaluated under our protocol, Shots=02026.06 | 39.5 | |
| Llama 3-8BParadigm=AR, Training Tokens=15T, Training Data=Not Released, Evaluation protocol=Reported by prior work, Shots=02026.06 | 39.2 | |
| LLaDA-8BParadigm=Masked Diffusion, Training Tokens=2.3T, Training Data=Not Released, Evaluation protocol=Reported by prior work, Shots=02026.06 | 38.7 | |
| Falcon-7BParadigm=AR, Training Tokens=1.5T, Training Data=Partially Released, Evaluation protocol=Evaluated under our protocol, Shots=02026.06 | 38.3 | |
| OLMo-7BParadigm=AR, Training Tokens=2.5T, Training Data=Fully Released, Evaluation protocol=Evaluated under our protocol, Shots=02026.06 | 37.9 | |
| Prot2Chat2026.02 | 21.58 | |
| ProtT32026.02 | 3.85 |