Vertex AI Model Garden is your one-stop-shop to over 100 cutting edge LLMs and task-specific models, from Google's own Palm 2 to open-source models like Llama 2 and Stable Diffusion.

Even if you've never deployed or managed a model before, Vertex makes it simple for any developer to find, deploy, and maintain foundation models.

In this session, we'll explore:
- Key features of Vertex AI Model Garden
- How to find the best model for your use case
- How to deploy your model to an endpoint you can easily integrate into your applications

03:54 Google's first party foundation models
07:34 Introduction to Vertex AI Model Garden
12:58 Tune models to fit your use case
20:28 Demo walkthrough of Llama 2 on Google Cloud
32:21 Managing your models with MLOps
38:30 Multi-task models and prompting
39:47 Selecting pre-trained models and prompting
41:30 Customization with tuning and curated data
44:30 Managing new artifacts
46:26 Evaluating and monitoring generated output
48:59 Connecting to Enterprise data
54:04 Three building blocks to support the Gen AI evaluation service
56:59 Safety bias evaluation
57:30 Summarization of building your MLOps strategy for Generative AI
1:00:24 Q&A

Resources:
Generative AI for Developers Learning Path: [ Ссылка ]
Explore and Evaluate Models using Model Garden: [ Ссылка ]
Join, learn, and engage with the Google Cloud Community: [ Ссылка ]

свернуть