Skip to content

Releases: Lin-jun-xiang/docGPT-langchain

v3.3

02 Oct 07:01
Compare
Choose a tag to compare
  • add .docx file loader
  • add .csv file loader
  • add .txt file loader

v3.2

25 Sep 08:24
Compare
Choose a tag to compare
  • add show available providers button
  • Update README

v3.1

29 Aug 08:07
Compare
Choose a tag to compare

✨What's new in version3?

  • Introduction of integration with gpt4free, allowing users to use docGPT for free without needing to input API keys or make payments.
  • If you choose to use the gpt4free model, you only need to select the Provider (default is g4f.provider.ChatgptAi). For more details about gpt4free, refer to the source project.

  • Version 2:

    • Uses the openai model.
    • Requires an openai_api_key. You can obtain this key from the link.
    • If you have a serpapi_key, AI responses can include Google search results.
  • Version 3:

    • Retains all the features of Version 2.
    • Introduces the gpt4free model, enabling completely free usage.
    • Users can choose between gpt4free and openai models:
      • gpt4free: Allows free access to OpenAI models through reverse engineering, but stability might be compromised.
      • openai: Offers stable access by using an API key.

We introduced the usage of gpt4free, allowing users to use the application for free without entering any API key or making payments.

  1. Add the gpt4free model for app, user can choose use gpt4free or openai model.

  2. In the gpt4free model, we'll use HuggingFace embedding for document which is no need openai api key.

  3. Clean and refactor code.

v2

24 Aug 07:14
Compare
Choose a tag to compare
v2
  1. Clean up app.py for cleaner code.

  2. Package the functionality of "Checking if the API key is entered and whether it is usable" into check_api_key.py.

  3. Use the streamlit_chat avatar parameter to add avatars to the chat bot. Ref

v1

23 Aug 08:14
Compare
Choose a tag to compare
v1
  1. The agent utility toolkit includes: search, calculator, and docGPT. The removal of LLM aims to allow the model to be more focused on document-related tasks.

  2. The implementation of docGPT involves FAISS + RetrievalQA. In the past, ChromaDB was used, but due to the issue mentioned here: GitHub Link, where there was a compatibility mismatch between ChromaDB and pysqlite3 versions causing the app to crash, it has been replaced with FAISS vector database.

  3. For langchain and streamlit, versions 0.0.218 and 1.22.0 are temporarily being used. Attempts were made with versions 0.0.228 and 1.24.1, but the model's responses were not satisfactory (reasons unknown).