Subscribe

ChatGLM-6B

Small bilingual Chinese-English chat model that runs on a consumer GPU; one of China's earliest widely used open LLMs.

A 6B chat model derived from the GLM-130B line, tuned for dialogue and runnable on consumer GPUs with INT4 quantization. It preceded Llama 2 (2023-07) and drew heavy community use; weights were free for research, with commercial use after registration.

Date
Tuesday 14 March 2023
Lab
Zhipu AI / Z.ai
Kind
open-weights
Access
open weights (restricted license)

Figures

MeasureValueMeasured by
GitHub stars (repo THUDM/ChatGLM-6B)40.9k
snapshot read 2026-10
independent

The GitHub repository was created on March 13, 2023 UTC, or March 14, 2023 in Beijing. Secondary reports give March 14, 2023 as the release date. Commercial use requires registration.

Sources

  1. github.com/THUDM/ChatGLM-6B
  2. arxiv.org/abs/2406.12793

This record was checked against its sources on 6 October 2026. How we check