News
ReleasesOVHcloud unveils Olala mini, its first in-house LLMChinaA 9B model beats a 1T model on FinQA, trained with Crusoe Managed RLReleasesThe internal metaphysics of an LLMReleasesMapping Language Model Internal Representations to Physical Laws and Scientific ReasoningResearchCan China keep rogue AI in check? Meet the start-up testing the top modelsFrontierMicrosoft aims to challenge OpenAI, Anthropic with cheaper, specialized AI modelsOpen modelsMicrosoft rolls out new first-party models and updates to agentic AI continuous improvementFrontierMicrosoft unveils AI models for coding, cybersecurity as it takes on Anthropic, OpenAIFrontierGoogle is making it easier to opt out of Gemini AI model training – here’s howFrontierOpenAI Unveils New Astra AI Model After Scrapping Prior ReleaseReleasesOVHcloud unveils Olala mini, its first in-house LLMChinaA 9B model beats a 1T model on FinQA, trained with Crusoe Managed RLReleasesThe internal metaphysics of an LLMReleasesMapping Language Model Internal Representations to Physical Laws and Scientific ReasoningResearchCan China keep rogue AI in check? Meet the start-up testing the top modelsFrontierMicrosoft aims to challenge OpenAI, Anthropic with cheaper, specialized AI modelsOpen modelsMicrosoft rolls out new first-party models and updates to agentic AI continuous improvementFrontierMicrosoft unveils AI models for coding, cybersecurity as it takes on Anthropic, OpenAIFrontierGoogle is making it easier to opt out of Gemini AI model training – here’s howFrontierOpenAI Unveils New Astra AI Model After Scrapping Prior Release
EveryLLM
ChatGLM-6B

ChatGLM-6B

Open bilingual dialogue model with 6.2 billion parameters

ChatGLM-6B is an open-source language model developed by the GLM team, featuring 6.2 billion parameters. It supports Chinese and English dialogue, runs on consumer hardware via quantization, and allows free commercial use after registration.

Share on WhatsApp

  • Contains 6.2 billion parameters
  • Trained on 1 trillion tokens
  • Runs on 6GB GPU memory with INT4
  • Free for commercial use after registration
  • Optimized for Chinese and English

What it is

ChatGLM-6B is a bilingual dialogue language model with 6.2 billion parameters. Built on the General Language Model (GLM) architecture, it is optimized for Chinese and English question answering. The model generates responses aligned with human preferences through supervised fine-tuning and reinforcement learning. It is designed to be lightweight enough for local deployment on standard consumer graphics cards.

How it came about

The model belongs to the GLM family, which includes larger variants like GLM-130B. ChatGLM-6B was released as an accessible version of the technology powering the ChatGLM service. It has since been succeeded by ChatGLM2-6B, which offers improved performance and longer context windows while retaining the low deployment barriers of the original.

Inside the model

The architecture uses the GLM framework. Training involved approximately one trillion tokens from Chinese and English corpora. The developers applied supervised fine-tuning, feedback bootstrap, and reinforcement learning with human feedback to align outputs. A key design feature is support for model quantization, allowing the INT4 version to run with only six gigabytes of GPU memory.

What it can do

The model handles bilingual dialogue and question answering. It aims to produce answers that match human preferences. By using quantization techniques, it maintains functionality on hardware with limited resources. The system manages conversation history to provide coherent multi-turn responses in both Chinese and English.

How to use it

Weights are fully open for academic research. Commercial use is permitted without fee after users complete a registration questionnaire. The codebase is licensed under Apache-2.0, while the model weights follow a specific Model License. Developers can access the model via Hugging Face and GitHub repositories.

Why it matters

ChatGLM-6B demonstrated that capable bilingual dialogue models could run on consumer-grade hardware. Its open licensing approach encouraged widespread adoption in both academic and commercial settings. It paved the way for subsequent iterations like ChatGLM2-6B within the GLM ecosystem.

Questions people ask about ChatGLM-6B

Can I use ChatGLM-6B for commercial projects?

Yes, free commercial use is allowed. You must complete a registration questionnaire provided by the developers to comply with the license terms.

What hardware do I need to run this model?

You can run it on consumer-grade graphics cards. Using INT4 quantization, the model requires only six gigabytes of GPU memory for local deployment.

Which languages does ChatGLM-6B support?

The model is optimized for bilingual use, specifically supporting Chinese and English. It is trained on a large corpus of both languages.

Good reading and sources

    Where this page comes from

    There is no English Wikipedia article on ChatGLM-6B yet, so the facts on this page come from the sources listed above: huggingface.co/zai-org/chatglm-6b. Some are the subject's own pages, so its claims about itself are reported as its claims. EveryLLM wrote the text in its own words, with AI assistance, and compares it with the sources to make sure no passage is copied.

    It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.

    The weekly digest

    Every Monday: the model of the day, new pages on EveryLLM, and the conferences and anniversaries of the week ahead.