Skip to content
Percona Software for MongoDB Documentation
Rate this page
☆ ☆ ☆ ☆ ☆
Thanks for your feedback
Thank you! The feedback has been submitted.

Get free database assistance or contact our experts for personalized support.

Get help from Percona

Percona Search for MongoDB 1.70.4-2 (2026-09-03)

Installation

Technical Preview

Percona Search for MongoDB 1.70.4-2 is available as a technical preview.

We recommend that early adopters use this release for testing purposes only and not in production environments.

We are pleased to introduce Percona Search for MongoDB 1.70.4-2. This release expands automatic embedding support with OpenAI-compatible providers, giving you more flexibility to use local or hosted embedding services with Percona Search for MongoDB.

This release of Percona Search for MongoDB is based on mongot 1.70.4, bringing the latest upstream improvements and fixes. For a complete list of changes, see the MongoDB release notes .

Release highlights

Automatic embedding with OpenAI-compatible providers

Percona Search for MongoDB now supports automatic embedding with services that implement the OpenAI /v1/embeddings API. You are no longer limited to Voyage AI. The new OPENAI_COMPATIBLE provider lets you use local or hosted embedding servers, including self-hosted options that can run without API keys or usage-based token charges.

This covers Ollama, vLLM, llama.cpp, LM Studio, LocalAI, and Hugging Face Text Embeddings Inference, alongside OpenAI and Azure OpenAI. Voyage AI and OpenAI-compatible models can be configured on the same mongot instance. The model specified in an autoEmbed index determines which embedding provider mongot uses.

With an autoEmbed Vector Search index, mongot updates embeddings for your documents as they change and creates an embedding for a query text at search time, so there is no additional pipeline to build or maintain. This extends upstream MongoDB Automated Embedding , which supports Voyage AI only.

Note

The OPENAI_COMPATIBLE provider currently supports float vector output only. Scalar and binary output formats aren’t supported.

For setup instructions and supported configurations, see our comprehensive documentation.

Changelog

New Features

  • PS4M-8: Percona Search for MongoDB now supports automatic embedding with services that implement the OpenAI /v1/embeddings API.

Join us in shaping the future of Percona Search for MongoDB

This technical preview is just the beginning. Your real-world experience and feedback are invaluable as we evolve Percona Search for MongoDB. We encourage you to explore it in your non-production testing and evaluation environments and share your insights:

  • What challenges did it solve for you?
  • What features would you like to see next?
  • How can we make it even better for your specific use cases?

We’d love to hear from you. If you have questions or feedback, contact your Percona representative or connect with the community on the Percona Community Forum.