# Liquid Extract VLM Judge leaderboard

Multimodal & Grounded benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/multimodal-grounded-liquid-extract-vlm-judge

Liquid image-to-JSON extraction VLM judge score. Every model the catalog carries a published Liquid Extract VLM Judge value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Multimodal & Grounded |
| Measure | VLM-judged extraction accuracy |
| Tasks | Image-to-JSON extraction |
| Difficulty | Structured visual extraction |
| Published by | [LiquidAI LFM2.5-VL Extract model cards](https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B-Extract) |

A display-only Liquid AI extraction metric measuring judged agreement between extracted values and the source image.

## Liquid Extract VLM Judge ranking

2 models with a published Liquid Extract VLM Judge value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | VLM-judged extraction accuracy | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [LFM2.5-VL-1.6B-Extract](https://modelchoice.dev/models/lfm2-5-vl-1-6b-extract) | LiquidAI | 90.6 | Observed 2026-11-21 · source benchlm:models |
| 2 | [LFM2.5-VL-450M-Extract](https://modelchoice.dev/models/lfm2-5-vl-450m-extract) | LiquidAI | 84.5 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
