unlimited-ocr-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| UNLIMITED_OCR_MODEL | No | Model name or path (transformers backend) | baidu/Unlimited-OCR |
| UNLIMITED_OCR_DEVICE | No | Device (transformers only) | cuda |
| UNLIMITED_OCR_BACKEND | No | Backend: 'transformers' or 'http' | transformers |
| UNLIMITED_OCR_SERVER_URL | No | vLLM/SGLang server URL (http backend) | http://127.0.0.1:10000 |
| UNLIMITED_OCR_TORCH_DTYPE | No | Torch dtype (transformers only) | bfloat16 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| ocr_parse_imageB | Parse a single image using Unlimited-OCR. Supports 'gundam' (crop mode) and 'base' modes. |
| ocr_parse_multiB | Parse multiple images (multi-page) using Unlimited-OCR. |
| ocr_parse_pdfC | Parse a PDF document using Unlimited-OCR. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
Each tool targets a distinct input type: a single image, multiple images, or a PDF. The descriptions make the boundaries clear, so an agent can select the correct tool without confusion.
All tools follow the consistent 'ocr_parse_<input>' pattern using snake_case. The naming is predictable and immediately communicates each tool's purpose.
Three tools is a well-scoped size for an OCR-focused server, covering the main input types without unnecessary bloat.
The tool surface covers the core OCR workflows: single images, multi-image batches, and PDFs. There are no obvious missing operations for the stated purpose.