Selecting and Switching Models in a Session

RecurseChat allows you to seamlessly switch between different AI models within the same chat session. This feature enables you to leverage the unique strengths of different models for various parts of your conversation.

Switching a model for the next conversation

1. Start a chat session with any model
2. Click the model name at the top of the chat window
3. Select a different model from the dropdown menu
4. Continue chatting - your next message will use the newly selected model

The chat history maintains continuity across model switches, allowing each model to see the full context of the conversation.

Regenerate Using Another Model

You can regenerate any previous response using a different model:

1. Go to a chat response
2. Click the regenerate icon
3. Select a different model from the dropdown
4. The response will be regenerated using the selected model

This allows you to:

• Compare how different models handle the same prompt
• Get alternative perspectives on complex questions
• Retry responses that weren’t quite what you needed
• Test model performance on specific types of queries

Supported Model Combinations

You can freely mix and match between:

• Local AI models (Llama, Mistral, etc.)
• OpenAI models (with API key)
• OpenAI-compatible endpoint models, including Ollama
• Anthropic Claude models (with API key)
• Custom GGUF models

Tips

• Each model maintains its own system prompt and settings
• You can switch models as many times as needed within a session
• When you start a new chat, the model selection will default to the last model you used