| agents-on-automationbench-public | LFM2.5-24B-A2B | – | Success Rate: 27.69 |
| agents-on-bfclv4 | LFM2.5-24B-A2B | – | Accuracy: 65.01 |
| agents-on-pinchbench | LFM2.5-24B-A2B | – | Accuracy: 83.18 |
| code-generation-on-scicode | LFM2.5-24B-A2B | – | Accuracy: 32.71 |
| code-generation-on-swe-bench-verified | LFM2.5-24B-A2B (OpenCode) | – | Accuracy: 51.00 |
| instruction-following-on-ifbench | LFM2.5-24B-A2B | – | Accuracy: 74.69 |
| instruction-following-on-multi-if | LFM2.5-24B-A2B | – | Accuracy: 84.38 |
| mathematical-reasoning-on-math-500 | LFM2.5-24B-A2B | – | Accuracy: 96.04 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | LFM2.5-24B-A2B | – | Accuracy: 74.42 |
| question-answering-on-humanity-s-last-exam-hle | LFM2.5-24B-A2B | – | Accuracy: 7.40 |