search

llm-tools / embedJs

A NodeJS RAG framework to easily work with LLMs and embeddings
https://llm-tools.mintlify.app/get-started/introduction
Apache License 2.0
333 stars 40 forks source link

Addition of an API for custom GPT actions #146

Open leVoT8 opened 4 weeks ago

leVoT8 commented 4 weeks ago

Discussed in https://github.com/llm-tools/embedJs/discussions/142

Originally posted by **leVoT8** October 26, 2024 ChatGPT provides paid subscribers with the opportunity to create custom GPTs to exist on their web user interface, enhancing usability and personalization. Through the use of "actions," users can tailor their GPTs to make API calls based on OpenAI's schema. To successfully implement a RAG model in a custom GPT, one must develop middleware that can handle the API call, transform the query into an embedding, process it against a vector database, and deliver the results back to the GPT. I believe EmbedJS has immense potential to play this middleware role. What is needed is an integrated API service that aligns with OpenAI's schema, complemented by comprehensive documentation to guide users in integrating this API schema into their custom GPT's actions. If we can bring this to fruition, I am confident that EmbedJS will experience a tremendous surge in popularity! The demand for such a solution is undeniable, and currently, no one has successfully filled this gap to my knowledge. For more information on GPT actions, check out the following resources: - [https://platform.openai.com/docs/actions/introduction](url) - [https://platform.openai.com/docs/actions/getting-started](url) - [https://chatgpt.com/g/g-TYEliDU6A-actionsgpt](url)