Code Generation on MLGym-Bench
75Pass@3MTL (I)
Evaluation Results
| Method | Links | |
|---|---|---|
| MTL (I)Model=GPT-5-mini, Method=MTL (I)2026.04 | 75 | |
| Zero-shotModel=GPT-5-mini, Method=Zero-shot2026.04 | 66.7 | |
| MTL (S)Model=GPT-5-mini, Method=MTL (S)2026.04 | 66.7 | |
| MTL (I)Model=DeepSeek V3.2, Method=MTL (I)2026.04 | 66.7 | |
| MTL (T)Model=GPT-5-mini, Method=MTL (T)2026.04 | 58.3 | |
| MTL (W)Model=GPT-5-mini, Method=MTL (W)2026.04 | 58.3 | |
| Zero-shotModel=DeepSeek V3.2, Method=Zero-shot2026.04 | 58.3 | |
| Zero-shotModel=Qwen3-Coder-480B-A35B-Instruct, Method=Zero-shot2026.04 | 58.3 | |
| MTL (I)Model=Qwen3-Coder-480B-A35B-Instruct, Method=MTL (I)2026.04 | 58.3 |