| code-generation-on-terminal-bench-2-0-terminus-2 | Nanbeige4.2-3B (Terminus-2) | #7 | Accuracy: 44.1 |
| mathematical-reasoning-on-hmmt-feb-2026 | Nanbeige4.2-3B | #7 | Accuracy: 82.8 |
| code-generation-on-scicode | Nanbeige4.2-3B | #18 | Accuracy: 35.6 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Nanbeige4.2-3B | #22 | Accuracy: 87.4 |
| code-generation-on-livecodebench-v6 | Nanbeige4.2-3B | #29 | Pass@1: 72.5 |
| question-answering-on-humanity-s-last-exam-hle | Nanbeige4.2-3B (w/o search) | #33 | Accuracy: 17.8 |
| instruction-following-on-ifbench | Nanbeige4.2-3B | #34 | Accuracy: 54.6 |
| code-generation-on-swe-bench-pro | Nanbeige4.2-3B | #37 | Resolved: 46.9 |
| code-generation-on-swe-bench-verified | Nanbeige4.2-3B (OpenHands) | #57 | Accuracy: 63.6 |