Local LLM (lllm)

A series of experiments with local LLMs

TODOs

Exp (Using Ollama)

• Run mistral 7b locally
• Run llama 2 locally
• Run custom prompts with llama 2
• Run API calls against llama/mistral

Exp (Get Twinny to work in VSCode)

• Done with caveats

Exp (Using Ollama for a Chat UI)

• Create chat UI similar to this

Exp (Create your own version of an open source LLM)

• Custom prompts + temps

Exp (Train Mistral on your own data - finetuning or use RAG)