LFM2.5-24B-A2B ranks first on ToolSandbox.
Open weights Flagship
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| IFEval | 96.8% | #4 of 126 | Model card | |
| Multi-IF | 84.4% | #3 of 28 | Model card | |
| IFBench | 74.7% | #24 of 70 | Model card | |
| Humanity's Last Exam | 7.40% | #141 of 147 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 74.4% | #143 of 284 | Model card | |
| MATH-500 | 96% | #34 of 67 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 51% | #124 of 150 | Model card | |
| SciCode | 32.7% | #42 of 47 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ToolSandbox | 80.7% | #1 of 5 | Model card | |
| LiveClawBench | 48.1% | #1 of 2 | Model card | |
| BFCL V4 | 65% | #13 of 24 | Model card | |
| AutomationBench | 27.7% | #29 of 41 | Model card | |
| PinchBench | 83.2% | #10 of 13 | Model card | |
| τ³-bench Banking | 8.04% | #19 of 20 | Model card | |
| WebArena-Verified | 26.8% | #6 of 6 | Model card |
The benchmarks the lab chose to headline, with the scores it gave.