Pinecone Documents API adds Gemini Embedding 2 at $0.20 per 1M tokens

Pinecone announced on November 12, 2026 that its Documents API now supports Google's Gemini Embedding 2. The API creates embeddings on writes and search queries through Vertex AI. The managed option costs $0.20 per 1 million tokens, as reported by Unite.AI.

What changes for developers

The managed option removes the need for a separate Google Cloud account, API key or embedding pipeline. Applications send plain text to Pinecone. Pinecone then invokes the model on every write and every query, and retries failed calls.

According to Unite.AI, the integration changes where the embedding work sits: it moves from the application to Pinecone's Documents API.

Key details

• Announced: November 12, 2026
• Embedding model: Google's Gemini Embedding 2
• Access path: Vertex AI, managed by Pinecone
• Price: $0.20 per 1 million tokens
• Requirements removed: separate Google Cloud account, API key and embedding pipeline
• Failure handling: Pinecone retries failed model calls

What this means

For teams building document search on Pinecone, embedding generation is now part of the Documents API rather than a separate service to operate. Pinecone handles the model call on both writes and queries, so the application only passes plain text. Teams that previously needed their own Google Cloud account, credentials and embedding code to use Gemini Embedding 2 no longer need them on this path. The $0.20 per 1 million tokens price gives teams a single figure to use when estimating embedding costs.