# Unlimited-OCR: Self-Hosting Hardware, Scenarios and Commercial License

> Compact document-vision model for one-shot long-document OCR, multi-page PDF parsing and structured layout extraction.

- Verified: 2026-09-12
- Released: 2026-06-22
- Canonical: https://chinaapi.ai/open-model-deployment/unlimited-ocr/
- Back to directory: https://chinaapi.ai/open-model-deployment/
- Machine-readable dataset: https://chinaapi.ai/data/open-model-deployment.json

## Model facts

- Vendor: Baidu
- Parameters: 3B total; 3B dense active
- Native precision: BF16
- Context: 32K tokens
- Category: Document OCR and vision-language models
- Modalities: image, PDF, text prompt → text, layout structure
- Serving paths: Transformers, vLLM, SGLang
- Official weights: https://huggingface.co/baidu/Unlimited-OCR?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment
- Model documentation: https://huggingface.co/baidu/Unlimited-OCR?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment

## Deployment evidence

- **Official launch / minimum — evidence C:** The official Transformers example loads the model on one CUDA GPU, but the vendor does not publish an absolute minimum VRAM configuration. Source: https://huggingface.co/baidu/Unlimited-OCR?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment
- **200-person team — evidence E:** Size by pages, pixels, document length and peak jobs rather than seats; start with one measured queue worker.
- **Commercial API — evidence C:** Official vLLM and SGLang serving paths exist; production needs replicated workers, document limits, processor pinning and post-processing observability. Source: https://huggingface.co/baidu/Unlimited-OCR?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment

## Scenario evidence

- **Long-document OCR — Vendor-stated:** Baidu positions the model for one-shot long-horizon OCR rather than page-fragment-only extraction. Source: https://huggingface.co/baidu/Unlimited-OCR?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment
- **Multi-page PDF parsing — Vendor-stated:** The official processor and examples cover document images and PDF-oriented workflows with a 32K generation configuration. Source: https://huggingface.co/baidu/Unlimited-OCR?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment
- **Structured layout extraction API — ChinaAPI inference:** The official vLLM and SGLang OpenAI-compatible paths make queued document extraction plausible after processor and post-processing validation.

## Commercial-use check

- License: MIT — https://huggingface.co/baidu/Unlimited-OCR/blob/main/LICENSE?utm_source=chinaapi&utm_medium=research&utm_campaign=open-model-deployment
- MaaS / hosted service: No model-specific MaaS restriction identified in the MIT license.
- Attribution: Retain the copyright and permission notice in copies or substantial portions of the weights.

## Avoid or validate first

- Sending unbounded PDFs without page, pixel and token admission limits
- Assuming OCR benchmark scores cover every language, handwriting or layout

## Known limitations and open questions

- No ChinaAPI hardware or OCR-quality reproduction
- PDFs still require deterministic image conversion and document post-processing
- The 32K setting is taken from the official inference configuration, not a full-context concurrency result

## Evidence boundary

Loading weights, completing a first forward pass and meeting a production latency/SLA are separate thresholds. The 200-person and commercial tiers require workload-specific measurement. License summaries are product research, not legal advice.

Generated by GENERATED BY scripts/gen_open_model_deployment.py.