LFM3-1.7B: benchmark results
Provider: Liquid AI. Released 2026-10-27. Access: Open.
Aggregate Rating 1584 ± 37, rank #501 of 1635 rated models, from 40 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LFM3-1.7B Release - AA-Omniscience (public) | 0.5 | Omniscience net score | 100 |
| LFM3-1.7B Release - AA-Omniscience non-hallucination (public) | 92.93 | Score (%) | 100 |
| LFM3-1.7B Release - AutomationBench (public) | 41.29 | Score (%) | 100 |
| LFM3-1.7B Release - BFCL V4 | 71.39 | Score (%) | 100 |
| LFM3-1.7B Release - Claw-Eval General (EN), average | 68.64 | Score (%) | 100 |
| LFM3-1.7B Release - IFBench | 73.03 | Score (%) | 100 |
| LFM3-1.7B Release - IFEval | 96.66 | Score (%) | 100 |
| LFM3-1.7B Release - IFStruct | 93.72 | Score (%) | 100 |
| LFM3-1.7B Release - LiveClaw | 45.6 | Score (%) | 100 |
| LFM3-1.7B Release - LiveCodeBench v6 | 73.46 | Score (%) | 100 |
| LFM3-1.7B Release - Multi-IF | 89.1 | Score (%) | 100 |
| LFM3-1.7B Release - ToolSandbox | 80.36 | Score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=lfm3-1-7b · How It Works · Data refreshed daily, snapshot 2026-11-05.