Prince Canuma comments

Results 572 comments of


                                            Prince Canuma

Enhancements to tool use in `gr.Chatbot`

My pleasure! Parallel calls is pretty much clear if we solve the first two.

Enhancements to tool use in `gr.Chatbot`

I saw it, but it was not very informative on how to do it. I have my custom library for inferencing Models locally on Apple Silicon that follows the openAI...

se_extractor.py crashes on line 28: "segments, info = model.transcribe(audio_path, beam_size=5, word_timestamps=True)"

Try changing the whisper model from `medium` to `tiny`. It worked for me :)

Expose custom filename, file prefix, output folder, audio format in server for audiobook generation

Thanks @rampadc, this is a great addition!

Expose custom filename, file prefix, output folder, audio format in server for audiobook generation

Please run pre-commit :)

Expose custom filename, file prefix, output folder, audio format in server for audiobook generation

@ivanfioravanti could use your help to double check this PR :)

Expose custom filename, file prefix, output folder, audio format in server for audiobook generation

Closing because #153 fixed

PaliGemma 2 mix segment multiple objects

> I am having trouble segmenting multiple objects when using PaliGemma 2 mix ("mlx-community/paligemma2-3b-mix-448-bf16", "mlx-community/paligemma2-10b-mix-448-8bit"). I also tried to directly use transformers and with the 3B model I sometimes get...

PaliGemma 2 mix segment multiple objects

If you could share the transformers examples as well would be nice Preferably with the images

PaliGemma 2 mix segment multiple objects

Yes, the problem of models like this and some OCR models like DeepSeek-OCR is that prompt matters. And for such tasks it's best to use bf16 or fp16, quants struggle...