# zai-org/GLM-OCR

GLM-OCR: Accurate ×  Fast × Comprehensive

Repository: https://github.com/zai-org/GLM-OCR
Canonical: https://ross.abutalabs.com/products/glm-ocr
Language: Python
License: Apache-2.0
License Family: permissive
Topics: glm, image2text, ocr
Last push: 2026-04-21T08:52:11+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 78, release rhythm 78, longevity 15
- inputs: {"age_days": 212, "days_push": 134, "days_rel": 147, "gap_med": 2.5, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7366, forks 664 (observed 2026-08-28T04:09:58.522654+00:00)

## What it is
GLM-OCR is an open-source 0.9B-parameter multimodal OCR model built on the GLM-V encoder-decoder architecture for complex document understanding, including tables, formulas, and code-heavy layouts. It ships with a Python SDK, CLI, and inference toolchain supporting deployment via vLLM, SGLang, and Ollama.

## Use cases
- extract text from scanned pdf documents
- recognize tables in images
- convert formulas in papers to latex
- parse complex document layouts into structured text
- run ocr at scale on a gpu server
- digitize receipts and sealed documents

## When to choose
- you need state-of-the-art document OCR with layout analysis
- you want a small efficient model deployable with vLLM or Ollama
- your documents contain tables, formulas, seals, or code

## When to avoid
- you only need simple plain-text OCR without layout understanding
- you have no GPU and cannot use the hosted API
- you need handwriting recognition in many languages

## Facets
- artifact type: library
- maturity: active
- function: ocr, machine-learning, llm-inference, image-processing, pdf
- domain: computer-vision, artificial-intelligence, pdf, deep-learning
- platform: python, cross-platform, cli
- tags: multimodal, document-understanding, vision-language-model, table-recognition, formula-recognition, vllm, sglang, ollama, layout-analysis, natural-language-processing, gpu

## Member repositories
- zai-org/GLM-OCR (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:58.522654+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:38:24.469499+00:00, confidence not recorded.
  - readme: https://github.com/zai-org/GLM-OCR (fetched 2026-08-28T04:09:58.522654+00:00, sha b6a03f7d431f)
- Data as of 2026-08-30T08:39:29.467469+00:00.