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 135 | Lab model card | |
| Multi-IF | 84.4% | #3 of 28 | Lab model card | |
| IFBench | 74.7% | #25 of 80 | Lab model card | |
| AA-Omniscience | -24.9 | #9 of 10 | Lab model card | |
| Humanity's Last Exam | 7.40% | #147 of 154 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH-500 | 96% | #34 of 70 | Lab model card | |
| GPQA Diamond | 74.4% | #152 of 302 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 51% | #127 of 155 | Lab model card | |
| SciCode | 32.7% | #42 of 49 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ToolSandbox | 80.7% | #1 of 5 | Lab model card | |
| LiveClawBench | 48.1% | #1 of 2 | Lab model card | |
| BFCL V4 | 65% | #16 of 30 | Lab model card | |
| PinchBench | 83.2% | #10 of 14 | Lab model card | |
| AutomationBench | 27.7% | #35 of 47 | Lab model card | |
| τ³-bench Banking | 8.04% | #21 of 22 | Lab model card | |
| WebArena-Verified | 26.8% | #7 of 7 | Lab model card |
The Self-Reported Index fits one score to the 12 capability results Liquid AI reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 134.1 to 140.4. How we compute it.
The benchmarks the lab chose to headline, with the scores it gave.