LangChain + Local LLMs + dstack

In this talk, you'll learn two things:

1. How and why to use langchain with local LLMs
2. How to use dstack to streamline ML development in the cloud

As we build a chat app (using the Falcon-7B-Instruct LLM).

Agenda

0. Poll.
1. What is LangChain?
2. Local LLMs with LangChain
3. Dev environments with dstack
4. Using Falcon with LangChain
5. Building a chat app
6. Automating tasks with dstack
7. Q&A

To work with this repo, make sure to install the latest version of dstack:

pip install "dstack[aws,gcp,azure]" -U

If you have any questions about dstack, please ask them in Slack.

References

• Pinecone's LangChain Handbook
• Hugging Face Local Pipelines
• Falcon-7B-Instruct
• Dolly 2.0's instruct pipeline
• dstack Docs
• Source code