**Guide to Using Hugging Face‘s ChatGLM**

Hugging Face has introduced the ChatGLM3-6B, the latest open-source model in the ChatGLM series. This guide will walk you through the features, dependencies, code usage, licensing, and citations related to ChatGLM3-6B.

**Introduction**:
ChatGLM3-6B is designed with a powerful base model, more comprehensive function support, and a more extensive open-source series. It aims to offer improved performance and utility in various chatbot applications.

**Dependencies**:
Before using ChatGLM3-6B, make sure to install the necessary dependencies using the following command:
“`shell
pip install protobuf transformers==4.30.2 cpm_kernels torch>=2.0 gradio mdtex2html sentencepiece accelerate
“`

**Code Usage**:
Once the dependencies are installed, you can generate dialogue using the following Python code:
“`python
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained(“THUDM/chatglm3-6b”, trust_remote_code=True)
model = AutoModel.from_pretrained(“THUDM/chatglm3-6b”, trust_remote_code=True).half().cuda()
model = model.eval()

response, history = model.chat(tokenizer, “你好”, history=[])
print(response)

response, history = model.chat(tokenizer, “晚上睡不着应该怎么办”, history=history)
print(response)
“`

For more detailed usage instructions, including running CLI and web demos, and model quantization, refer to the [Github repository](https://github.com/THUDM/ChatGLM).

**License**:
The code for ChatGLM3-6B is open-sourced under the Apache-2.0 license. However, the use of the ChatGLM3-6B model weights requires compliance with the Model License.

**Citation**:
If you find ChatGLM3-6B helpful, consider citing the following papers:
“`
@article{zeng2022glm,
title={Glm-130b: An open bilingual pre-trained model},
author={Zeng, Aohan and Liu, Xiao and Du, Zhengxiao and Wang, Zihan and Lai, Hanyu and Ding, Ming and Yang, Zhuoyi and Xu, Yifan and Zheng, Wendi and Xia, Xiao and others},
journal={arXiv preprint arXiv:2210.02414},
year={2022}
}

@inproceedings{du2022glm,
title={GLM: General Language Model Pretraining with Autoregressive Blank Infilling},
author={Du, Zhengxiao and Qian, Yujie and Liu, Xiao and Ding, Ming and Qiu, Jiezhong and Yang, Zhilin and Tang, Jie},
booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
pages={320–335},
year={2022}
}
“`

Utilize this guide to maximize your experience with Hugging Face‘s ChatGLM3-6B for your chatbot applications.

Source link
## Hugging Face ChatGLM3-6B Manual

Welcome to the Hugging Face ChatGLM3-6B manual! In this guide, you will learn how to use the ChatGLM3-6B, a powerful open-source model for generating dialogue. Whether you are a developer, researcher, or someone interested in natural language processing, this manual will provide you with the information you need to get started with ChatGLM3-6B.

### Introduction
ChatGLM3-6B is the latest open-source model in the ChatGLM series. It introduces powerful features such as a more robust base model, comprehensive function support, and a wider open-source series. If you want to experience the larger-scale ChatGLM model, you can visit [chatglm.cn](https://www.chatglm.cn).

### Dependencies
Before getting started with ChatGLM3-6B, you need to ensure that you have the required software dependencies installed. You can install the dependencies using the following command:
“`shell
pip install protobuf transformers==4.30.2 cpm_kernels torch>=2.0 gradio mdtex2html sentencepiece accelerate
“`

### Code Usage
To generate dialogue using the ChatGLM3-6B model, you can use the following code:
“`python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained(“THUDM/chatglm3-6b”, trust_remote_code=True)
model = AutoModel.from_pretrained(“THUDM/chatglm3-6b”, trust_remote_code=True).half().cuda()
model = model.eval()
response, history = model.chat(tokenizer, “你好”, history=[])
print(response)
response, history = model.chat(tokenizer, “晚上睡不着应该怎么办”, history=history)
print(response)
“`
For more detailed usage instructions, including running CLI and web demos, and model quantization, please refer to the [Github Repo](https://github.com/THUDM/ChatGLM).

### License
The code in the ChatGLM3-6B repository is open-sourced under the [Apache-2.0 license](https://huggingface.co/THUDM/chatglm3-6b/blob/main/LICENSE). The use of the ChatGLM3-6B model weights needs to comply with the [Model License](https://huggingface.co/THUDM/chatglm3-6b/blob/main/MODEL_LICENSE).

### Citation
If you find the ChatGLM3-6B model helpful, please consider citing the following papers:
“`bib
@article{zeng2022glm,
title={Glm-130b: An open bilingual pre-trained model},
author={Zeng, Aohan and Liu, Xiao and Du, Zhengxiao and others},
journal={arXiv preprint arXiv:2210.02414},
year={2022}
}

@inproceedings{du2022glm,
title={GLM: General Language Model Pretraining with Autoregressive Blank Infilling},
author={Du, Zhengxiao and Qian, Yujie and Liu, Xiao and others},
booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
pages={320–335},
year={2022}
}
“`

Thank you for using the Hugging Face ChatGLM3-6B model! If you have any further questions or need assistance, please refer to the [Github Repo](https://github.com/THUDM/ChatGLM) for support.

💻 Github Repo • 🐦 Twitter • 📃 [GLM@ACL 22] [GitHub] • 📃 [GLM-130B@ICLR 23] [GitHub]

👋 Join our Slack and WeChat

📍Experience the larger-scale ChatGLM model at chatglm.cn



介绍 (Introduction)

ChatGLM3-6B 是 ChatGLM 系列最新一代的开源模型,在保留了前两代模型对话流畅、部署门槛低等众多优秀特性的基础上,ChatGLM3-6B 引入了如下特性:

  1. 更强大的基础模型: ChatGLM3-6B 的基础模型 ChatGLM3-6B-Base 采用了更多样的训练数据、更充分的训练步数和更合理的训练策略。在语义、数学、推理、代码、知识等不同角度的数据集上测评显示,ChatGLM3-6B-Base 具有在 10B 以下的预训练模型中最强的性能。
  2. 更完整的功能支持: ChatGLM3-6B 采用了全新设计的 Prompt 格式,除正常的多轮对话外。同时原生支持工具调用(Function Call)、代码执行(Code Interpreter)和 Agent 任务等复杂场景。
  3. 更全面的开源序列: 除了对话模型 ChatGLM3-6B 外,还开源了基础模型 ChatGLM-6B-Base、长文本对话模型 ChatGLM3-6B-32K。以上所有权重对学术研究完全开放,在填写问卷进行登记后亦允许免费商业使用

ChatGLM3-6B is the latest open-source model in the ChatGLM series. While retaining many excellent features such as smooth dialogue and low deployment threshold from the previous two generations, ChatGLM3-6B introduces the following features:

  1. More Powerful Base Model: The base model of ChatGLM3-6B, ChatGLM3-6B-Base, employs a more diverse training dataset, more sufficient training steps, and a more reasonable training strategy. Evaluations on datasets such as semantics, mathematics, reasoning, code, knowledge, etc., show that ChatGLM3-6B-Base has the strongest performance among pre-trained models under 10B.
  2. More Comprehensive Function Support: ChatGLM3-6B adopts a newly designed Prompt format, in addition to the normal multi-turn dialogue. It also natively supports function call, code interpreter, and complex scenarios such as agent tasks.
  3. More Comprehensive Open-source Series: In addition to the dialogue model ChatGLM3-6B, the base model ChatGLM-6B-Base and the long-text dialogue model ChatGLM3-6B-32K are also open-sourced. All the weights are fully open for academic research, and after completing the questionnaire registration, they are also allowed for free commercial use.



软件依赖 (Dependencies)

pip install protobuf transformers==4.30.2 cpm_kernels torch>=2.0 gradio mdtex2html sentencepiece accelerate



代码调用 (Code Usage)

可以通过如下代码调用 ChatGLM3-6B 模型来生成对话:

You can generate dialogue by invoking the ChatGLM3-6B model with the following code:

>>> from transformers import AutoTokenizer, AutoModel
>>> tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True)
>>> model = AutoModel.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True).half().cuda()
>>> model = model.eval()
>>> response, history = model.chat(tokenizer, "你好", history=[])
>>> print(response)
你好👋!我是人工智能助手 ChatGLM-6B,很高兴见到你,欢迎问我任何问题。
>>> response, history = model.chat(tokenizer, "晚上睡不着应该怎么办", history=history)
>>> print(response)
晚上睡不着可能会让你感到焦虑或不舒服,但以下是一些可以帮助你入睡的方法:

1. 制定规律的睡眠时间表:保持规律的睡眠时间表可以帮助你建立健康的睡眠习惯,使你更容易入睡。尽量在每天的相同时间上床,并在同一时间起床。
2. 创造一个舒适的睡眠环境:确保睡眠环境舒适,安静,黑暗且温度适宜。可以使用舒适的床上用品,并保持房间通风。
3. 放松身心:在睡前做些放松的活动,例如泡个热水澡,听些轻柔的音乐,阅读一些有趣的书籍等,有助于缓解紧张和焦虑,使你更容易入睡。
4. 避免饮用含有咖啡因的饮料:咖啡因是一种刺激性物质,会影响你的睡眠质量。尽量避免在睡前饮用含有咖啡因的饮料,例如咖啡,茶和可乐。
5. 避免在床上做与睡眠无关的事情:在床上做些与睡眠无关的事情,例如看电影,玩游戏或工作等,可能会干扰你的睡眠。
6. 尝试呼吸技巧:深呼吸是一种放松技巧,可以帮助你缓解紧张和焦虑,使你更容易入睡。试着慢慢吸气,保持几秒钟,然后缓慢呼气。

如果这些方法无法帮助你入睡,你可以考虑咨询医生或睡眠专家,寻求进一步的建议。

关于更多的使用说明,包括如何运行命令行和网页版本的 DEMO,以及使用模型量化以节省显存,请参考我们的 Github Repo

For more instructions, including how to run CLI and web demos, and model quantization, please refer to our Github Repo.



协议 (License)

本仓库的代码依照 Apache-2.0 协议开源,ChatGLM3-6B 模型的权重的使用则需要遵循 Model License

The code in this repository is open-sourced under the Apache-2.0 license, while the use of the ChatGLM3-6B model weights needs to comply with the Model License.



引用 (Citation)

如果你觉得我们的工作有帮助的话,请考虑引用下列论文。

If you find our work helpful, please consider citing the following papers.

@article{zeng2022glm,
  title={Glm-130b: An open bilingual pre-trained model},
  author={Zeng, Aohan and Liu, Xiao and Du, Zhengxiao and Wang, Zihan and Lai, Hanyu and Ding, Ming and Yang, Zhuoyi and Xu, Yifan and Zheng, Wendi and Xia, Xiao and others},
  journal={arXiv preprint arXiv:2210.02414},
  year={2022}
}
@inproceedings{du2022glm,
  title={GLM: General Language Model Pretraining with Autoregressive Blank Infilling},
  author={Du, Zhengxiao and Qian, Yujie and Liu, Xiao and Ding, Ming and Qiu, Jiezhong and Yang, Zhilin and Tang, Jie},
  booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages={320--335},
  year={2022}
}


The

tag in HTML is a flexible and powerful tool that can be used for various purposes. Here are some potential use cases for the

tag in the context of artificial intelligence, frameworks, Python, coding, Hugging Face, creation, AI, Flutter, Dialogflow, Firebase, Google Cloud, and database applications.

1. Structure and Layout: The

tag is commonly used to create the structural layout of a web page. In the context of AI, frameworks, Python, and coding, it can be used to organize different sections of a website or application, such as the header, footer, sidebar, and main content area. This can be particularly useful when developing AI-powered web applications or dashboards.

2. Data Display and Visualization: In the context of Hugging Face and AI frameworks, the

tag can be used to display and visualize data. This can include the presentation of AI model outputs, training performance metrics, and other relevant information. The

tag can be styled and customized using CSS to create visually appealing and informative displays.

3. User Interface Components: When building AI-powered applications using frameworks like Flutter, the

tag can be used to create various user interface components such as buttons, dropdowns, input fields, and interactive elements. These components can be linked to AI functionalities and provide users with seamless interactions with intelligent systems.

4. Dialogflow Integration: In the context of Dialogflow and Firebase, the

tag can be used to integrate chatbot interfaces within web applications. This allows users to interact with AI-powered chatbots for customer support, information retrieval, and other AI-related tasks. The

tag can be used to define the chat interface and handle dynamic content updates.

5. Database Integration: In the context of Google Cloud and database applications, the

tag can be used to present data retrieved from databases. This can include displaying query results, visualizing data trends, and creating interactive data reports. The

tag can be utilized to structure and render database-driven content on web pages.

6. Content Creation and Editing: The

tag can be used as a container for creating and editing rich content within web applications. This can be particularly relevant in the context of content creation tools and AI-powered content generation platforms. The

tag can be used to create dynamic text editors and multimedia content creation interfaces.

Overall, the

tag is a versatile and essential element in HTML that can be leveraged for a wide range of use cases in the context of artificial intelligence, frameworks, Python, coding, Hugging Face, creation, AI, Flutter, Dialogflow, Firebase, Google Cloud, and database applications. Its flexibility and customization options make it an indispensable tool for web development and AI integration.

2024-01-20T07:48:28+01:00