| Benchmark | Model | Rank | Results |
|---|---|---|---|
| instruction-following-on-multi-if | LFM2.5-2.6B | #1 | Accuracy: 80.07 |
| agents-on-pinchbench | LFM2.5-2.6B | #7 | Accuracy: 68.22 |
| agents-on-bfclv4 | LFM2.5-2.6B | #9 | Accuracy: 56.88 |
| deep-research-agents-on-browsecomp-plus | LFM2.5-2.6B (OpenClaw) | #17 | Accuracy: 26.89 |
| instruction-following-on-ifbench | LFM2.5-2.6B | #28 | Accuracy: 59.17 |
| code-generation-on-livecodebench-v6 | LFM2.5-2.6B | #42 | Pass@1: 59.41 |
| mathematical-reasoning-on-aime-2025 | LFM2.5-2.6B | #83 | Accuracy: 51.87 |