LFM2.5-24B-A2B: benchmark results
Provider: Liquid AI. Released 2026-10-13. Access: Open.
Aggregate Rating 1631 ± 26, rank #742 of 10916 rated models, from 24 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Liquid LFM2.5 24B Card - AutomationBench | 27.69 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - BFCLv4 | 65.01 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - IFBench | 74.69 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - IFEval | 96.75 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - LiveClaw | 48.09 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - Multi-IF | 84.38 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - ToolSandbox | 80.71 | Score (%) | 100 |
| Liquid LFM2.5 24B Card - τ³-Bench Banking | 8.04 | Score (%) | 100 |
| ToolSandbox | 80.71 | Avg Score (imported) | 100 |
| Multi-IF | 84.38 | Average (imported) | 77.8 |
| Tinybird AI SQL Benchmark - Exactness | 50.77 | Result exactness vs human reference queries (0-100) | 64 |
| ReasonScape R12 | 831.25 | ReasonScore | 54.1 |
Interactive version: theaggregate.ai/model?slug=lfm2-5-24b-a2b · How It Works · Data refreshed daily, snapshot 2026-11-08.