Liquid image-to-JSON extraction schema consistency F1 (Liquid Extract F1)
We show this table for reference; we do not rank on it.
A display-only Liquid AI extraction metric measuring field-name agreement between requested schema fields and extracted JSON fields.
Benchmark score on Liquid Extract F1 — November 13, 2026
We compile the Liquid Extract F1 rows from provider self-reports. LFM2.5-VL-1.6B-Extract leads the table at 99.6%, followed by LFM2.5-VL-450M-Extract (98.8%). We do not use these results to rank models overall.
LFM2.5-VL-1.6B-Extract
LiquidAI
Reported by LiquidAILFM2.5-VL-450M-Extract
LiquidAI
Reported by LiquidAI2 modelsMultimodal & GroundedCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (2 models)
All 2 results on this page were reported by the labs that built the models.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | LFM2.5-VL-1.6B-ExtractLiquidAIReported by LiquidAI | 99.6% | Reported by LiquidAI | Not reported | Open |
| 2 | LFM2.5-VL-450M-ExtractLiquidAIReported by LiquidAI | 98.8% | Reported by LiquidAI | Not reported | Open |
- Lab-reported
- Benchmark-run
- Independent
About Liquid Extract F1
Year
2026
Tasks
Image-to-JSON extraction
Format
Schema field F1
Difficulty
Structured visual extraction
Liquid reports Schema Consistency F1 as a macro-averaged set-level F1 over predicted versus requested field names. BenchLM keeps it separate from general OCR and VQA metrics.
Freshness and provenance
Version
Liquid Extract F1 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.
Questions
What does Liquid Extract F1 measure?
A display-only Liquid AI extraction metric measuring field-name agreement between requested schema fields and extracted JSON fields.
Which model scores highest on Liquid Extract F1?
LFM2.5-VL-1.6B-Extract by LiquidAI currently leads with a score of 99.6% on Liquid Extract F1.
How many models are evaluated on Liquid Extract F1?
2 AI models have published results on Liquid Extract F1 in the BenchLM catalog.
Compare top models on Liquid Extract F1
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.