Allow different model for Recorder post-processing

[ChrisBBBB]

I use the Recorder function and have a second prompt for after-transcription processing. It hadn't occurred to me that the model it uses for that second processing is whatever the model is set for the brain/chat. So when I've been experimenting with DeepSeek, all my transcriptions were taking much, much longer to come back to me. And that's because the post-processing was being sent to DeepSeek which was completely overthinking it.

In some future version, would it be possible to allow us to set different models for the recorder function? We can already set which transcription model it goes to, but if we could also set which post-processing model it goes to please (and maybe some of its basic model settings like temperature too). For my use case, I would want a fast, cheap LLM for the post-processing, but use something much more complex for my chat model.

Thanks.

[mentatbot]

If you would like me to solve this issue, either tag me in a comment or check this box:

• Solve Issue

_{You can disable automatic comments on my settings page}