Multimodal · Benchmark profile
A display-only Liquid AI extraction metric measuring judged agreement between extracted values and the source image.
Data verified 21 Jul 2026 · Methodology 1.6.0
Benchmark score on Liquid image-to-JSON extraction VLM judge score
LFM2.5-VL-1.6B-Extract leads at 90.6%, followed by LFM2.5-VL-450M-Extract (84.5%).
Visual analysis
Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.
2 ranked models · higher is better
| Model | Rank | Provider | Score |
|---|---|---|---|
| LFM2.5-VL-1.6B-Extract | #1 | LiquidAI | 90.6% |
| LFM2.5-VL-450M-Extract | #2 | LiquidAI | 84.5% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | LFM2.5-VL-1.6B-Extract lfm2-5-vl-1-6b-extract | LiquidAI | open | Estimated reference | 90.6% |
| #2 | LFM2.5-VL-450M-Extract lfm2-5-vl-450m-extract | LiquidAI | open | Estimated reference | 84.5% |
About Liquid image-to-JSON extraction VLM judge score
A display-only Liquid AI extraction metric measuring judged agreement between extracted values and the source image. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A display-only Liquid AI extraction metric measuring judged agreement between extracted values and the source image.
LFM2.5-VL-1.6B-Extract by LiquidAI currently leads with 90.6%.
2 models in the LuminaBench cohort have a qualifying score on this benchmark.
Related
