# zai-org/GLM-130B

GLM-130B: An Open Bilingual Pre-Trained Model (ICLR 2023)

Repository: https://github.com/zai-org/GLM-130B
Canonical: https://ross.abutalabs.com/products/glm-130b
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2023-07-25T09:01:49+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1491, "days_push": 1135, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7651, forks 599 (observed 2026-08-28T04:10:02.186420+00:00)

## What it is
GLM-130B is an open bilingual (English and Chinese) 130-billion-parameter dense language model pre-trained with the General Language Model (GLM) algorithm, released with code and checkpoints for reproducible evaluation. It supports inference on a single A100 or V100 server, with INT4 quantization reducing requirements to 4x RTX 3090 GPUs.

## Use cases
- run inference with a 130B bilingual English-Chinese language model
- reproduce benchmark results for large language models on 30+ tasks
- quantize a large model to INT4 to fit on consumer GPUs
- evaluate zero-shot performance on Chinese CLUE and FewCLUE benchmarks
- compare against GPT-3, OPT, and BLOOM on LAMBADA and MMLU

## When to choose
- you need an open bilingual Chinese-English large model with downloadable checkpoints
- you want reproducible evaluation of a 130B model on limited hardware via INT4 quantization
- you need strong zero-shot Chinese language understanding performance

## When to avoid
- you need a modern chat model with long context - use ChatGLM2-6B or newer instead
- you only need English-only models or smaller deployable models
- you lack multi-GPU server hardware and cannot use quantization

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, llm-training, machine-learning, nlp
- domain: large-language-models, artificial-intelligence, deep-learning
- platform: python
- tags: bilingual, chinese-english, glm, pretrained-model, quantization, int4, iclr-2023, model-checkpoints, natural-language-processing, linux, gpu, docker

## Member repositories
- zai-org/GLM-130B (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:02.186420+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 2026-08-29T17:36:45.673801+00:00, confidence not recorded.
  - readme: https://github.com/zai-org/GLM-130B (fetched 2026-08-28T04:10:02.186420+00:00, sha 67572ddba381)
- Data as of 2026-08-30T08:39:29.467469+00:00.