Open-weight structured extraction models
28 in the catalogue today. Every one with what it costs, who sells it and where it stands.
Tools that pull structure out of unstructured input — fields from an invoice, a schema from a page, a table from a report. Some are models, some are services with a model inside. Judge them on what they do when the input is malformed, because that is the whole job; anything can parse a clean file.
Weights published under a licence that lets you run them where you like and sell what you build — Apache 2.0, MIT and their kin impose little beyond keeping the notice. This is the list to start from if you need the model on your own hardware, in your own region, or simply want no third party between you and it. The trade is that hosting is now your problem, and the prices shown beside each one are what somebody else charges to do it for you.
- AutoGLM-Phone-9B-MultilingualZ.ai$0.25→$1per Mtok in / out3 selling
- BiomedCLIP-PubMedBERT_256-vit_base_patch16_224Microsoft
- DPT-2LandingAI$0.03per page1 selling
- DeepScaleR-1.5B-PreviewAgentica
- DeepSeek V3.2 ExpDeepSeek1st of 16· 18 boards$0.21→$0.32per Mtok in / out8 selling
- DeepSeek V4 ProDeepSeek26th of 290· 29 boards$1.32→$3.96per Mtok in / out21 selling
- DeepSeek v3.2DeepSeek1st of 47· 34 boards$0.21→$0.31per Mtok in / out19 selling
- Document AIMistral AI$0.005per page1 selling
- DolphinByteDance
- GLM-OCRZ.ai$0.03→$0.03per Mtok in / out1 selling
- K2-Horizon-MoVA-36B-A4BInstitute of Foundation Models90th of 302
- LongCat 2.0Meituan LongCat122nd of 302· 3 boards$0.75→$2.95per Mtok in / out3 selling
- MinerU2.5-Pro-2604-1.2BOpenDataLab
- MiniCPM-SALAOpenBMB
- NuExtractNuMind
- NuExtract-1.5NuMind
- NuExtract3NuMind
- OmniParserMicrosoft
- Qianfan-OCRBaidu$0.067→$0.27per Mtok in / out1 selling
- Qwen3.5 9BAlibaba125th of 313· 12 boards$0.08→$0.13per Mtok in / out8 selling
- Qwen3.5-0.8BAlibaba261st of 302$0.01→$0.05per Mtok in / out1 selling
- Qwen3.5-122B-A10BAlibaba21st of 262· 11 boards$0.22→$2.2per Mtok in / out8 selling
- Qwen3.5-35B-A3B-BaseAlibaba$0.54→$1.33per Mtok in / out, training lane1 selling
- Qwen3.6 27BAlibaba60th of 290· 14 boards$0.3→$2per Mtok in / out12 selling
- Step 3.7 FlashStepFun123rd of 302· 14 boards$0.2→$1.2per Mtok in / out9 selling
- TildeOpen-30bTildeAI
- dots.ocrdots studio
- needle2Cactus Compute