Prince Canuma comments

Results 572 comments of


                                            Prince Canuma

Issue with Finetuning Mistral (Mistral-Small-3.1-24B-Instruct-2503-4bit)

We are refactoring the trainer Try this branch #261

Issue with Finetuning Mistral (Mistral-Small-3.1-24B-Instruct-2503-4bit)

cc: @Goekdeniz-Guelmez

fix: (outetts)loading model: Speaker file not found.

Thanks @zysam! We will merge as soon as the tests clear

Add ESpeakNG.xcframework for iOS on Kokoro MLX Audio Swift

Thank you very much @rudrankriyam! I agree with you @lucasnewman but from my initial testing this is the fastest way to unlock IOS. Since this is only for Kokoro. What...

Add ESpeakNG.xcframework for iOS on Kokoro MLX Audio Swift

Thanks @niklasmato this is great feedback! Could you please try out this PR and open an issue with some examples if you find this the same limitations.

Add ESpeakNG.xcframework for iOS on Kokoro MLX Audio Swift

> Thank you very much @rudrankriyam! > > I agree with you @lucasnewman but from my initial testing this is the fastest way to unlock IOS. > > Since this...

Add ESpeakNG.xcframework for iOS on Kokoro MLX Audio Swift

@rudrankriyam could you resolve the conflicts?

Excessive Memory Allocation (56.98GB) on M2 Mac Mini (32GB) Using Qwen2-VL-2B-Instruct-4bit

Hey @WildKratts Could you share a reproducible example? I know the current trainer has a few bugs to iron out.

Excessive Memory Allocation (56.98GB) on M2 Mac Mini (32GB) Using Qwen2-VL-2B-Instruct-4bit

Closing stale

Apply PR #319 fixes to Qwen 2.5VL position id

Thank you very much @ddupont808! Small nit: As suggested in @prncvrm's PR please move the position ID calculations inside the language.py it can be exactly in the LanguageModel() otherwise it...