glm-ocr:bf16

glm-ocr:bf16

566 Downloads Updated 8 hours ago

GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.

vision

ollama run glm-ocr:bf16

curl http://localhost:11434/api/chat \
  -d '{
    "model": "glm-ocr:bf16",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='glm-ocr:bf16',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'glm-ocr:bf16',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 8 hours ago

8 hours ago

6effedd0dc8a · 2.2GB ·

model

archglmocr

·

parameters1.11B

·

quantizationF16

2.2GB

params

{ "temperature": 0 }

18B

Readme

Note: this model requires Ollama 0.15.5, which is currently being published as a pre-release version.

GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture. The model integrates the CogViT visual encoder pre-trained on large-scale image–text data, a lightweight cross-modal connector with efficient token downsampling, and a GLM-0.5B language decoder.

Usage

Text recognition

ollama run glm-ocr Text Recognition: ./image.png

Table recognition

ollama run glm-ocr Table Recognition: ./image.png

Figure recognition

ollama run glm-ocr Figure Recognition: ./image.png

Key features

State-of-the-Art Performance: Achieves a score of 94.62 on OmniDocBench V1.5, ranking #1 overall, and delivers state-of-the-art results across major document understanding benchmarks, including formula recognition, table recognition, and information extraction.
Optimized for Real-World Scenarios: Designed and optimized for practical business use cases, maintaining robust performance on complex tables, code-heavy documents, seals, and other challenging real-world layouts.
Efficient Inference: With only 0.9B parameters, GLM-OCR supports deployment via vLLM, SGLang, and Ollama, significantly reducing inference latency and compute cost, making it ideal for high-concurrency services and edge deployments.
Easy to Use: Fully open-sourced and equipped with a comprehensive SDK and inference toolchain, offering simple installation, one-line invocation, and smooth integration into existing production pipelines.

![logo.svg](/assets/library/glm-ocr/c08dfddc-0734-4983-a450-3466d01594ee)

> Note: this model requires Ollama 0.15.5, which is currently being published as a [pre-release version](http://github.com/ollama/ollama/releases/).

GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture. The model integrates the CogViT visual encoder pre-trained on large-scale image–text data, a lightweight cross-modal connector with efficient token downsampling, and a GLM-0.5B language decoder.

![image.png](/assets/library/glm-ocr/7b4bd5ed-e6bc-4f36-889d-a02fa36e88ae)

### Usage

#### Text recognition

```shell
ollama run glm-ocr Text Recognition: ./image.png
```

#### Table recognition

```shell
ollama run glm-ocr Table Recognition: ./image.png
```

#### Figure recognition

```shell
ollama run glm-ocr Figure Recognition: ./image.png
```

### Key features

- **State-of-the-Art Performance**: Achieves a score of 94.62 on OmniDocBench V1.5, ranking #1 overall, and delivers state-of-the-art results across major document understanding benchmarks, 
including formula recognition, table recognition, and information extraction.

- **Optimized for Real-World Scenarios**: Designed and optimized for practical business use cases, maintaining robust performance on complex tables, code-heavy documents, seals, and other challenging real-world layouts.

- **Efficient Inference**: With only 0.9B parameters, GLM-OCR supports deployment via vLLM, SGLang, and Ollama, significantly reducing inference latency and compute cost, making it ideal for high-concurrency services and edge deployments.

- **Easy to Use**: Fully open-sourced and equipped with a comprehensive [SDK](https://github.com/zai-org/GLM-OCR) and inference toolchain, offering simple installation, one-line invocation, and smooth integration into existing production pipelines.

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)