Skip to main content

Liquid image-to-JSON extraction VLM judge score (Liquid Extract VLM Judge)

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

Benchmark score on Liquid Extract VLM Judge — June 8, 2026

BenchLM mirrors the published score view for Liquid Extract VLM Judge. LFM2.5-VL-1.6B-Extract leads the public snapshot at 90.6% , followed by LFM2.5-VL-450M-Extract (84.5%). BenchLM does not use these results to rank models overall.

2 modelsMultimodal & GroundedCurrentDisplay onlyUpdated June 8, 2026

About Liquid Extract VLM Judge

Year

2026

Tasks

Image-to-JSON extraction

Format

VLM-judged extraction accuracy

Difficulty

Structured visual extraction

Liquid reports VLM Judge Score using a separate vision model to compare extracted JSON content against the image. BenchLM stores it as a specialized visual extraction quality signal.

BenchLM freshness & provenance

Version

Liquid Extract VLM Judge 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Benchmark score table (2 models)

1
90.6%
2
84.5%

FAQ

What does Liquid Extract VLM Judge measure?

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

Which model scores highest on Liquid Extract VLM Judge?

LFM2.5-VL-1.6B-Extract by LiquidAI currently leads with a score of 90.6% on Liquid Extract VLM Judge.

How many models are evaluated on Liquid Extract VLM Judge?

2 AI models have been evaluated on Liquid Extract VLM Judge on BenchLM.

Compare Top Models on Liquid Extract VLM Judge

Last updated: June 8, 2026 · BenchLM version Liquid Extract VLM Judge 2026

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.