LFM2 1.2B: benchmark results
Liquid AI's LFM2-1.2B, one of four LFM2 checkpoints from 350M to 2.6B parameters, built with a hybrid architecture of multiplicative gates and short convolutions for edge deployment. Provider: Liquid AI. Released 2025-07-10. Access: Open.
Aggregate Rating 1287 ± 17, rank #1503 of 1656 rated models, from 351 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Polish Summarization - PSC | 25.18 | chrF3++ (%) | 92.5 |
| Enkrypt AI - Bias Risk | 68.48 | Risk Score | 79.6 |
| EuroEval Spanish Summarization - Mlsum ES | 26.96 | chrF3++ (%) | 70.1 |
| EuroEval Portuguese Summarization - Publico | 32.53 | chrF3++ (%) | 68.5 |
| Enkrypt AI - Toxicity Risk | 2.88 | Risk Score | 56.6 |
| Kurakurai French LLM - MATH-500 | 40.94 | Accuracy (%) | 56.1 |
| Kurakurai French LLM - MGSM | 56.52 | Accuracy (%) | 56.1 |
| Kurakurai French LLM - MultiIF | 38.77 | Accuracy (%) | 56.1 |
| EuroEval Albanian Summarization - LR SUM SQ | 29.61 | chrF3++ (%) | 55.2 |
| Kurakurai French LLM - GPQA Diamond | 26.16 | Accuracy (%) | 53.7 |
| Kurakurai French LLM - IFEval | 59.53 | Accuracy (%) | 53.7 |
| EuroEval Serbian Summarization - LR SUM SR | 25.85 | chrF3++ (%) | 51 |
Interactive version: theaggregate.ai/model?slug=lfm2-1-2b · How It Works · Data refreshed daily, snapshot 2026-10-20.