Build generative apps faster with Vertex AI
summary
Announces six new Vertex APIs for production gen‑AI workflows: a Document Understanding API (DocAI‑based) that extracts sections, tables and structure; an improved Gecko embedding model that ranks highly on leaderboards; Vector Search with hybrid search and YouTube/Ads‑scale efficiency for large indexes; a Ranking API that scores retrieved passages per question; a Gemini‑fine‑tuned grounded generation with citations; and a Check Grounding API that verifies each sentence against evidence and flags contradictions. Integrates with LangChain and llama_index.
AI-written summary · about
chapters (10)
- 0:00Intro
- 1:49Document AI API
- 2:40Embedding API
- 2:59Vector search API
- 3:26Ranking API
- 3:58Grounded generation API
- 4:23Check grounding API
- 5:16Best features of these new APIs
- 7:16How can developers integrate these APIs?
- 8:14Wrap up
related talks
15mLatest Generative AI products and solutions on
32mGenerative AI Text and Multimodal Embedding Models for Real World Use Cases
31mBringing AI to Your Data
24mEmpowering Vue Apps with AI
23mScaling GenAI Applications and Agentic AI for Business Competitive Differentiation
49mEnter the Brave New World of GenAI with Vector Search
34mEnter the Brave New World of GenAI with Vector Search
17mThe keys to success with Generative AI and Vector Databases