Deploy Hugging Face models on Google Cloud: directly from Vertex AI

In this series of three videos, I walk you through the deployment of Hugging Face models on Google Cloud, in three different ways:
- Deployment from the hub model page to Inference endpoints ( • Deploy Hugging Face mo... ), with the Google Gemma 7B model,
- Deployment from the hub model page to Vertex AI ( • Deploy Hugging Face mo... ), with the Microsoft Phi-2 2.7B model,
- Deployment directly from within Vertex AI (this video), with the TinyLlama 1.1B model.
Get started at huggingface.co :)
⭐️⭐️⭐️ Don't forget to subscribe to be notified of future videos. Follow me on Medium at / julsimon or Substack at julsimon.substack.com. ⭐️⭐️⭐️

[@LindsayHiebert]

Great job! Thank you!!!

[@juliensimonfr]

Glad it was helpful!

[@LindsayHiebert]

Julien , excellent episodes. Do you have a cheat sheet reference guide for your video series to develop top level app development skills with Hugging Face models. I am looking for your recommendations for the best path to learn and practice with the best practices and tools. Thanks much!

[@juliensimonfr]

Hi Lindsay, if you're new to Hugging Face, I recommend starting at huggingface.co/learn, in particular the NLP course and the cookbook. Then, you can start diving deeper into whatever makes more sense to you: cloud, optimization, etc.

[@LindsayHiebert]

Thanks much Julien!

[@brightworld7550]

Very helpfull thank you so much. One question. When it comes to pricing. Does Vertex AI bill you based on the usage (how many seconds the model is running) or for as long as the model is running 24/7?

[@juliensimonfr]

It's instance-based, so you pay for instance time as long as it's up.

[@donkeroo1]

I see at 2:14 you have an error for the gemma model deployment, I am getting the same error for many of the models from HF, however I did manage to get tinyllama working. Any advice on chasing down the error? Logs are telling me the models already exist. Thank you for the guidance!

[@juliensimonfr]

Looks like the first few days were a bit rough :) Are you still seeing errors? If yes, please post at discuss.huggingface.co