Pinecone integrated embedding now supports Google's Gemini Embedding 2
Blog post from Pinecone
Pinecone’s integrated embedding feature for its Documents API now supports Google’s Gemini Embedding 2, allowing teams to perform semantic search by sending plain text without managing Google Cloud credentials, API calls, rate limits, retries, or separate billing. Pinecone generates embeddings for document writes and search queries through Google Vertex AI, bills tokens at Google’s list price, and can pair semantic and full-text search on the same text field. Users configure Gemini Embedding 2 when creating a serverless Documents API index, with support for adjustable embedding dimensions, automatic document and query instructions, and task-specific modes such as general search, classification, clustering, question answering, fact verification, and code retrieval. The model supports more than 100 languages but processes requests globally without a guaranteed region. It is available on Builder plans and above through the console and Python SDK 10.3.0, while Node.js and Go support is planned; it currently applies only to new text-based Documents API indexes, with no support yet for BYOC, pod-based, Records API, or multimodal indexes. Builder plans include 10 million monthly Gemini Embedding 2 tokens, temporarily doubled to 20 million through December 31, 2026.
| Trend | Post Mentions | Total Month Mentions | Posts | Companies | MoM |
|---|---|---|---|---|---|
| Vector Search | 37 | 1,251 | 223 | 92 | -37% |
| Serverless | 1 | 833 | 236 | 92 | -40% |
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.