A display-only Liquid AI extraction metric measuring the share of image-to-JSON outputs that parse as strict JSON.
BenchLM mirrors the published score view for Liquid Extract JSON Validity. LFM2.5-VL-1.6B-Extract leads the public snapshot at 99.6% , followed by LFM2.5-VL-450M-Extract (98.9%). BenchLM does not use these results to rank models overall.
LFM2.5-VL-1.6B-Extract
LiquidAI
LFM2.5-VL-450M-Extract
LiquidAI
Year
2026
Tasks
Image-to-JSON extraction
Format
Strict JSON parseability rate
Difficulty
Structured visual extraction
Liquid evaluates Extract models on 2,000 image, schema, and JSON triples. BenchLM stores JSON Validity as a specialized display-only extraction signal rather than a weighted multimodal benchmark.
Version
Liquid Extract JSON Validity 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
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.
A display-only Liquid AI extraction metric measuring the share of image-to-JSON outputs that parse as strict JSON.
LFM2.5-VL-1.6B-Extract by LiquidAI currently leads with a score of 99.6% on Liquid Extract JSON Validity.
2 AI models have been evaluated on Liquid Extract JSON Validity on BenchLM.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.