Agent Switchboard

Converts PDFs and images into Markdown, HTML or JSON with layout preserved, handling tables, forms, handwriting and 90+ languages, via a CLI with local or vLLM inference.

Use it now

Keep it with up to 9 other tools you're setting up, then share them all as one link.

Try this first

Ask your coding agent:

  • Layout-Preserving OCR

    Use Chandra to convert images and PDFs into Markdown, HTML or JSON: ‹what you need›

  • Local or vLLM Inference

    Use Chandra to run the model: ‹what you need›

Install the CLI

Install it as its own instructions say.

Its page: github.com/datalab-to/chandra

Call the API

Auth None needed

Its page: github.com/datalab-to/chandra

Compare

Press Compare to add it to the tray at the bottom of the window, then pick up to 2 others anywhere on the site.

Access
CLIAPI
Auth
None
Skills
3

Skills

  • Layout-Preserving OCR

    Converts images and PDFs into Markdown, HTML or JSON with detailed layout information, and extracts images and diagrams with captions.

  • Tables, Forms and Math

    Reconstructs forms including checkboxes, complex tables, math and handwriting across 90+ languages from scanned or digital pages.

  • Local or vLLM Inference

    Runs the model locally through HuggingFace with --method hf, or against a vLLM server launched with chandra_vllm, for single files or folders.

Data & AnalyticsInfrastructure & Opsocrpdf-parsingdocument-extractionlayout-analysishandwriting-recognitionvllmhuggingfacedatalab
Visit Agent
chandra-ocr
Converts PDFs and images into Markdown, HTML or JSON with layout preserved, handling tables, forms, handwriting and 90+ languages, via a CLI with local or vLLM inference.
try this first
Ask your coding agent:
Use Chandra to convert images and PDFs into Markdown, HTML or JSON: ‹what you need›layout-ocr
Use Chandra to run the model: ‹what you need›inference-modes
use it now
None neededapi.auth
fields
nameChandra
providerDatalab
urlhttps://github.com/datalab-to/chandra
categoriesdata-analytics · infrastructure
accesscli · api
authnone
streamingfalse
pushfalse
verifiedtrue
officialfalse
sponsoredfalse
tagsocr, pdf-parsing, document-extraction, layout-analysis, handwriting-recognition, vllm, huggingface, datalab
skills
layout-ocrLayout-Preserving OCRConverts images and PDFs into Markdown, HTML or JSON wi…
tables-forms-mathTables, Forms and MathReconstructs forms including checkboxes, complex tables…
inference-modesLocal or vLLM InferenceRuns the model locally through HuggingFace with --metho…
alternatives
/agents/chandra-ocr/alternatives8 listedpaddleocr, somark-oxr, mineru
compare
/compare?a=chandra-ocr&b=paddleocr&c=somark-oxrwith paddleocr, somark-oxr