LFM2.5-VL-3B: benchmark results
Provider: Liquid AI. Released 2026-08-12. Access: Open.
Aggregate Rating 1494 ± 33, rank #2448 of 10916 rated models, from 79 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LFM3-VL-2B Release - TextVQA (val) | 90.4 | Published normalized score (0–100) | 100 |
| LFM3-VL-2B Release - POPE | 88.9 | Published normalized score (0–100) | 80 |
| LFM3-VL-2B Release - RefCOCO-M | 94.4 | Published normalized score (0–100) | 80 |
| LFM3-VL-2B Release - RefCOCO-avg | 87.9 | Published normalized score (0–100) | 80 |
| LLM Stats (TextVQA) | 84.3 | Score (%) | 75 |
| LFM3-VL-2B Release - CV-Bench | 86.3 | Published normalized score (0–100) | 70 |
| LFM3-VL-2B Release - OCRBench v1 | 84.7 | Published normalized score (0–100) | 70 |
| LFM3-VL-2B Release - V*Bench | 81.2 | Published normalized score (0–100) | 70 |
| LFM3-VL-2B Release - ZoomBench (full + crop average) | 61.6 | Published normalized score (0–100) | 70 |
| LFM3-VL-2B Release - MMMB | 83 | Published normalized score (0–100) | 60 |
| LFM3-VL-2B Release - Point-Bench | 58.7 | Published normalized score (0–100) | 60 |
| LFM3-VL-2B Release - SEED-Bench (image) | 77.7 | Published normalized score (0–100) | 60 |
Interactive version: theaggregate.ai/model?slug=lfm2-5-vl-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.