Datalab Launches 9B Open-Weight Vision Model lift, Converting PDFs to Structured JSON
Decision Brief
What changedDatalab releases lift, a 9B open-weight vision model that converts PDFs and images into schema-compliant JSON.
Why it mattersAI builders should focus on such specialized open-source models to simplify document parsing and structured data extraction.
Who should careTeams building on model APIs
Affected stackNo specific stack identified
Builder actionMonitor
Source confidenceMedium · Reliable media or first-hand reporting
Datalab's lift model has 9B parameters and open weights, specializing in extracting structured JSON from PDFs and images. It uses schema-constrained decoding to ensure correct output format and is trained to return null for missing fields instead of hallucinating. In a benchmark with 225 documents, lift achieved 90.2% field accuracy.
Summary basis: official / RSS sourceUnless it says 'full article read', this summary is based only on publicly available content — it never pretends to have read restricted originals.
Sources
- MarkTechPost
Fast research-paper and ML tooling summaries, useful for infra and agent updates.
- MarkTechPost