# liucongg/ChatGLM-Finetuning

基于ChatGLM-6B、ChatGLM2-6B、ChatGLM3-6B模型，进行下游具体任务微调，涉及Freeze、Lora、P-tuning、全参微调等

Repository: https://github.com/liucongg/ChatGLM-Finetuning
Canonical: https://ross.abutalabs.com/products/chatglm-finetuning
Language: Python
License Family: other
Topics: chatglm, chatgpt, freeze, lora, p-tuning, chatglm2, chatglm3
Last push: 2023-12-12T15:36:52+00:00

## Health v2 (maintenance only)
Score: 21/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 89
- inputs: {"age_days": 1246, "days_push": 995, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2771, forks 308 (observed 2026-08-28T04:07:18.916436+00:00)

## What it is
A Python toolkit for fine-tuning ChatGLM-6B, ChatGLM2-6B, and ChatGLM3-6B large language models using Freeze, LoRA, P-Tuning, and full-parameter methods. It supports single- and multi-GPU training via DeepSpeed and includes evaluation for information extraction, generation, and classification tasks.

## Use cases
- fine-tune chatglm on custom instruction data
- compare lora vs p-tuning vs freeze fine-tuning results
- train chatglm3-6b with deepspeed on multiple gpus
- adapt chatglm for information extraction tasks
- fine-tune a chinese llm on a single gpu with limited vram
- run full-parameter fine-tuning of chatglm2-6b

## When to choose
- you specifically want to fine-tune ChatGLM-family models
- you want to experiment with and compare multiple PEFT methods (LoRA, P-Tuning, Freeze, full-parameter)
- you need DeepSpeed-based single- or multi-GPU training without the HuggingFace Trainer abstraction

## When to avoid
- you use models other than ChatGLM/ChatGLM2/ChatGLM3
- you need actively maintained code for the latest model releases
- you want a production-ready training framework with broad model support

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-training, machine-learning, deep-learning
- domain: large-language-models, machine-learning
- platform: python
- tags: chatglm, lora, p-tuning, freeze, deepspeed, fine-tuning, parameter-efficient-finetuning, chinese-llm, natural-language-processing, gpu, linux

## Member repositories
- liucongg/ChatGLM-Finetuning (main) score 21

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:18.916436+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 2026-08-30T08:16:39.534124+00:00, confidence not recorded.
  - readme: https://github.com/liucongg/ChatGLM-Finetuning (fetched 2026-08-28T04:07:18.916436+00:00, sha 066df352630e)
- Data as of 2026-08-30T08:39:29.467469+00:00.