LLM æčĄæĻ
å ¨éĸį大č¯č¨æ¨Ąååēåč¯åãæ¯čžå¨įŧį¨ãæ¨įååé æ§äģģåĄä¸į襨į°ã
| æ¨Ąå | ä¸ä¸æ | åšŗå°äģˇæ ŧ | åŽæšäģˇæ ŧ | GPQA | AIME 2025 | SWE-Bench | ARC-AGI v2 | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| æĒæžå°įŦĻåæĄäģļįæ¨Ąå | |||||||||||
ææ åŽäš
LLM
- GPQA
- Graduate-level science questions requiring expert knowledge.
- AIME 2025
- Recent math competition problems.
- SWE-Bench
- Real GitHub issues requiring code changes.
- ARC-AGI v2
- Abstract reasoning problems.