LFM2 1.2B — benchmark results
Provider: Liquid AI. Released 2025-11-28. Access: Open.
Unified ELO 1221 ± 16, rank #1661 of 1776 rated models, from 304 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Polish Summarization - PSC | 25.18 | Score (%) | 92.5 |
| EuroEval Spanish Summarization - Mlsum ES | 26.96 | Score (%) | 70.1 |
| EuroEval Portuguese Summarization - Publico | 32.53 | Score (%) | 68.5 |
| EuroEval Albanian Summarization - LR SUM SQ | 29.61 | Score (%) | 55.2 |
| EuroEval Serbian Summarization - LR SUM SR | 25.85 | Score (%) | 51 |
| EuroEval Bosnian Summarization - LR SUM BS | 25.77 | Score (%) | 49.3 |
| EuroEval French Summarization - Orange SUM | 35.17 | Score (%) | 49.3 |
| EuroEval German Summarization - Mlsum DE | 31.77 | Score (%) | 49.3 |
| EuroEval Catalan NLU - ScaLA CA | 2.98 | Linguistic acceptability Score (%) | 45.5 |
| EuroEval Catalan Summarization - Dacsa CA | 32.19 | Score (%) | 43.5 |
| AA Humanity's Last Exam | 5.7 | Accuracy (%) | 41.3 |
| EuroEval Lithuanian Common Sense Reasoning | 4.1 | Common Sense Reasoning Average Score (%) | 40.8 |
Interactive version: theaggregate.ai/model?slug=lfm2-1-2b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.