Compatible AI Models with Pieces
Pieces utilizes cloud-hosted LLMs from providers like OpenAI, Anthropic, and Google. All local models are served through Ollama, a core dependency of PiecesOS and the rest of the Pieces Suite.
Cloud-Only LLMs | Providers
Browse the list of cloud-hosted AI models available for use with the Pieces Desktop App and several of our plugins and extensions.
| Provider | Model Name |
|---|---|
| OpenAI | GPT-X |
| Anthropic | Claude / Sonnet / Opus / Haiku |
| Gemini / Pro / Flash / Chat |
Local-Only LLMs | Providers
Read through the list of local AI models available for use within the Pieces Desktop App and the rest of the Pieces Suite.
| Provider | Model Name |
|---|---|
| Gemma / Code Gemma | |
| IBM | Granite / Code / Dense / MoE |
| Meta | LLaMA / CodeLLaMA |
| Mistral | Mistral / Mixtral |
| Microsoft | Phi |
| Qwen | QwQ / Coder |
| StarCoder | StarCoder |
Using Ollama with Pieces
Ollama is required to utilize local generative AI features.
It’s a lightweight system that allows for a plug-and-play experience with local models, meaning that Pieces can update the number of compatible models we serve at lightning-speeds!
If you want to read more about installing and using Ollama with Pieces, click here for the Ollama section of our Core Dependencies documentation.
