**Last Update**: 25.05.2024 *** - [Model](https://ollama.com/library) link on Ollama site - llama3 (Meta) - phi3 (Microsoft) - mistral - mixtral (Mixture of Experts, based on mistral) - gemma (open source version of Gemini) - Vision models (interpret images) - llava - bakllava - moondream - Performance of models is highly dependent on *size* - Small: 7b or fewer - Medium: ~15b parameters - Large: ~26b to 70b parameters - Each model uses a *tag* to represent *size* *** **References**: - [Running Uncensored and Open Source LLMs on Your Local Machine](https://www.youtube.com/watch?v=Q0toFxwB_Is)