How to Use Hugging Face Inference API

This video is a hands-on step-by-step tutorial with code to show you how to use hugging face inference API locally for free.
▶ Become a Patron 🔥 - / fahdmirza
#huggingface
PLEASE FOLLOW ME:
▶ LinkedIn: / fahdmirza
▶ KZbin: / @fahdmirza
▶ Blog: www.fahdmirza.com
RELATED VIDEOS:
▶ Code www.fahdmirza....
▶ HF huggingface.co...
All rights reserved © 2021 Fahd Mirza

[@kij0te]

I am a beginner, after many hours of failed attempts with other KZbin videos, yours worked, THANK YOU, you are the best

[@fahdmirza]

Glad I could help!

[@Alvaro-cc6dt]

Thank you so much, my teacher asked me to make a similar tutorial and i based on yours, but dont worry i put your video as reference, greetings from Brazil

[@fahdmirza]

wow, so nice of you, thanks. Please also share the channel with your teacher and class. Thanks

[@DmitriZaitsev]

What was the exact code you used? The one you linked to produces another output. Thanks.

[@samratajamal]

I want to deploy openface model for my application but unable to do this and not even showing the deploy option.can you tell me how I deploy this model because The model you use can be deployed.

[@shubhamnikumbh4117]

How we can make sure the model we pick for repo id. At what quantization type? Because in many models comes with the quantization.

[@zanjeeltariq9079]

Thank you so much, God bless!!!!

[@fahdmirza]

You are so welcome

[@adityapatil6723]

sir i want to use "openai/whisper-large-v3-turbo" model but i cant find any tutorial on this

[@TestAutomationTV]

Why'd you use a giant model locally?

[@vigneshrs3221]

Can I use this api just like the open ai api? And maybe use for my projects?

[@adityapatil6723]

sir i admire for just making such a grate vid

[@fahdmirza]

Appreciate it!

[@NetZeroEarth]

Can u provide the code use in github or on your website please?

[@fahdmirza]

Sure, I have put the link to code in video's description.

[@NetZeroEarth]

Thank you !!! 😊

[@dallasdominguez9501]

Fahd great video! Now When you say we can work with any model on huggingface, does this also include models that accept multimodal inputs? Im on a project where I need to make calls to models like InstructBLIP and LLaVA-Next which are found in the huggingface transformers library. However I'm getting stuck on both how I would make the call to the actual model, how I would structure the prompt and json packet to send to the model (given that I need to send multiple images, data from a csv file, and how I'd be able to integrate all of this into a proper prompt that could be sent to a multimodal model). Any tips or guidance would help. Thank you!

[@fahdmirza]

would have to check.

[@nocode_fever]

Same question

[@hitthetone9635]

Im having difficulty with exporting the token. I think in Windows its done with "set" instead of "export", but it still doesnt accept this.

[@fahdmirza]

yes set or setx

[@tillwill3232]

Hi, does the free inference api work for custom/pivate models or do we need to use the paid endpoints? Thanks

[@fahdmirza]

I think there are some free credits available.

[@ketan6182]

very often when i try to run code this way it does not run at all, it just sits blankly waiting for output... is there a solution to this significant waiting time? microsoft/Phi-3-vision-128k-instruct is the model i was using.

[@saarza9991]

try running in parallel thread using async await

[@joshuamickey1648]

Hi Mirza, Can we do tool call through this HFInference API? Reply fast if possible.

[@fahdmirza]

yes you can if the model supports it.

[@merakshot9277]

i know you explained it but i want just to make sure how to know if a model is paid, or free, and does Inference API can be used for free like lifetime or it will stop one day

[@merakshot9277]

thank you for the video, and i will appreciate your reply

[@innovatorshome]

@@merakshot9277 you need to pay for the endpoint

[@dronnema203]

is that an ai agent??

[@fahdmirza]

No

[@loicnembot5927]

thanks