Switching LLMs

The Pieces for VS Code Extension currently supports 26 different LLMs (which includes both cloud-based and local models).


Available LLMs

We constantly update and configure our plugins and extensions, like the Pieces for VS Code Extension, to work with the latest LLMs.

See the list of available models below.

Cloud LLMs

The cloud-based models integrate effortlessly with the Pieces Copilot, offering high-performance and real-time responses to your queries.


GPT-4o Mini

Gemini-1.5 Pro

Claude-3 Sonnet

GPT-4o

Gemini-1.5 Flash

Claude-3.5 Sonnet

GPT-4 Turbo

Gemini Pro Chat

Code Chat Bison (PaLM 2)

GPT-4

Claude-3 Haiku

Chat Bison (PaLM 2)

GPT-3.5 Turbo

Claude-3 Opus

Custom OpenAI API Key


On-Device LLMs

We also support on-device LLMs for developers prioritizing privacy, security, or offline functionality.


Mistral 7B

LLama-2 7B

Code Gemma-1.1 7B

Phi-2

Llama-3 8B

Granite-8B

Phi-3 Mini 128K / Phi-3 Mini 4K

Gemma-1.1 7B / Gemma-1.1 2B

Granite-3B


How To Configure Your LLM Runtime

Switching your LLM model in the Pieces for VS Code Extension is a straightforward process, giving you the flexibility to choose the model that best suits your needs.

How to change your LLM:

  1. Open the Copilot Chat view by clicking the Pieces Copilot icon in the sidebar

  2. Located the Active Model in the bottom-left corner of the view where the current model (e.g., Claude 3.5 Sonnet) is displayed

  1. Click on Change Model to open the Manage Copilot Runtime modal

  2. Browse the list of local and cloud models, and select your preferred model

From here, you can browse and select from a variety of available models, such as the local and cloud-based models listed in the table at the top of this page.

Once you’ve chosen a new model, the switch is instant, allowing you to continue your work seamlessly with the selected model's capabilities—no need to restart or refresh anything.


Cloud-hosted models offer access to the latest AI capabilities, while on-device models ensure offline functionality, making Pieces Copilot adaptable to your specific workflow and environment.


Updated on