How Model Switcher works

The "model" refers to the AI model that MATE will run in the background to fulfill your request. MATE uses models from OpenAI and from Anthropic ("Claude"). More models will be supported in the future. Send your requests to support@omata.io

Differences

All models are excellent at what they do. The main differences seem to lie in their coding abilities. However, given the generative nature of these models, which one performs better really depends on your specific task. I encourage you to experiment and see for yourself what works best for you.

OpenAI

• gpt-4o: good for general requests and serves as a strong all-rounder.

• o3-mini: fast "reasoning" model that thinks before it acts (slower than gpt-4o)

Anthropic

• Claude 3.5 Sonnet: tends to perform better on complex coding tasks. If you’re struggling with scripts, it’s definitely worth giving Claude a try.

• Claude 3.7 Sonnet: officially the next version of 3.5. But it's more intelligent and some people say that it tends to do "too much". At least, it's very motivated :)

Switch model in a chat

Like with all prompt settings (at the bottom) you can switch the model for every message, if you like.
Click on the model switcher and it will show you a "drop-up" where you can set your model.
On hover it shows with a tooltip which model is active.
You can switch the model for every request.

Configure the default state for new chats

In Settings > Chats you define which model will be used as the default for new chats.
In each chat you can switch models again at any time.