Whether LangChain supports ChatGLM model?

[gao-stu]

Whether LangChain supports ChatGLM model? If yes, how to invoke the ChatGLM model? If no, how to access the customized model?
I want to use ChatGLM just like the code below uses openAI, getting an api_key to call ChatGLM

from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain

os.environ["OPENAI_API_KEY"] = "..."

llm = OpenAI(temperature=0.9)

prompt = PromptTemplate(
    input_variables=["product"],
    template="What is a good name for a company that makes {product}?",
)
print(prompt.format(product="colorful socks"))```

[shashankdeshpande]

@gao-stu
Yes, we can define custom LLM as mentioned here - How to define Custom LLM?

This code snippet might be helpful to integrate ChatGLM with Langchain - Langchai + ChatGLM

[gao-stu]

Thank you! It's very useful.I have successfully used a custom llm.
But,There is another question, how does langchain use the open-source embeding? I want to use my own embeding and llm

[shashankdeshpande]

Langchain offers multiple options for embeddings.

For example, if you prefer using open-source embeddings from huggingface or sentence-transformers, you can find more information at this link - HuggingFace Embeddings

Alternatively, if you prefer to create custom function for obtaining embeddings, this might be helpful - Fake Embeddings
You can integrate your custom function as _get_embedding method in above code snippet

Hope this helps!

[gao-stu]

Thanks a lot, this is very useful