WTF is going on?

#71
by vbarrier - opened

Since 2 weeks, this is endpoint is a mess! And no answer anywhere!

On the 28/04/2025, It is not working AT ALL : API says that the accept-type is incorrect.. (and obviously it's fine)

And now as of 29/04/2025 => 404

Guys?

What seems to be the problem Monsieur Vincent?

This dedicated Inference Endpoint is broken as of 7/18/25

Would appreciate a fix, this model is great

We moved on to voxtral. cheaper, support is there and works better.

We can get it Going

We discovered it works if you manually set the container to โ€œdefaultโ€

@erikkaum @michellehbn could you take a look when you get a chance (requested by Dmitry from OpenAI)?

Hi @vbarrier ๐Ÿ‘‹

Sorry to hear that things aren't working.
Just to double check: are you deploying this in the Inference Endpoints app or is are you using it through the inference providers? ๐Ÿค”

Hi everyone! Thanks for waiting while we took a deeper look. We've just applied a change so that this can be deployed easily from the model card, and from our model catalog in Inference Endpoints here: /static-proxy?url=https%3A%2F%2Fendpoints.huggingface.co%2Fcatalog%3Fquery%3Dwhisper%3C%2Fa%3E. Let us know though if you run into any issues and we'd love to know how it goes! ๐Ÿค—

Hi Guys, thanks for supporting and being with me on this journey. I am parsing things out. So I will be available. Cooper Ka**CK7 & 7 & KAiOS & Adapt

Iโ€™m going to hire a few people to help with a Microsoft project. Iโ€™ll make sure we upload some things. Anyone want to work on a project see where the next next step goes? I have ten days a month that I have not used, so that is the rough time commitment. Cooper K. 7, ck7 etc. !!//!!

Hi @vbarrier ๐Ÿ‘‹

Sorry to hear that things aren't working.
Just to double check: are you deploying this in the Inference Endpoints app or is are you using it through the inference providers? ๐Ÿค”

I was using inference providers

Gotcha, are you still facing issues or do you have an error that keeps occurring?

I just tested the example snippet and everything seemed to work but you might have hit an edge case or similar:

curl /static-proxy?url=https%3A%2F%2Frouter.huggingface.co%2Fhf-inference%2Fmodels%2Fopenai%2Fwhisper-large-v3-turbo \
    -X POST \
    -H "Authorization: Bearer $HF_TOKEN" \
    -H 'Content-Type: audio/flac' \
    --data-binary @"sample1.flac"

Happy to help ๐Ÿ‘

Sign up or log in to comment