AI
With the Hype around ChatGPT, good to gather some resources
Prompt Engineering
- Awesome ChatGPT Prompts
- Pre-prime ChatGTP
- You’re Using ChatGPT Wrong! Here’s How to Be Ahead of 99% of ChatGPT Users
- Why you (probably) Don’t Need to Fine Tune an LLM: Seems like advance prompt engineering might get you there.
Github Copilot
Copilot Alternatives
- HuggingFace VSCode Plugin. - leveraging the StarCoder model
- FauxPilot: Attempt for locally-hosted Copilot (nVidia GPUs required)
- LocalPilot: Another attempt for a locally-hosted Copilot
Alternative Models
- A List of Billion+ Parameter Models
- GPT4All: An ecosystem of open-source chatbots trained on a massive collections of clean assistant data including code, stories and dialogue
- OpenAssistant: Open source ChatAI
- Hello Dolly: Democratizing the magic of ChatGPT with open models
Local LLMs
- LocalGPT: Based on PrivateGPT, but replaced the GPT4ALL model with Vicuna-7B model and we are using the InstructorEmbeddings instead of LlamaEmbedding
- OnPrem.LLM: (although as of 9/8/2023, does NOT suport GGUF format models…)
- PrivateGPT: Ask question of your documentation
- Slowllama: llama2-70b tuned for Macbook Air M1/M2