-

Elasticsearch Open Inference API Extends Support for Hugging Face Models with Semantic Text

Applications using Hugging Face embeddings on Elasticsearch now benefit from native chunking

SAN FRANCISCO--(BUSINESS WIRE)--Elastic (NYSE: ESTC), the Search AI Company, today announced the Elasticsearch Open Inference API now supports Hugging Face models with native chunking through the integration of the semantic_text field. Developers can now quickly ship generative AI (GenAI) applications without the burden of writing custom chunking logic, leveraging the Elasticsearch Open Inference API integration with Hugging Face Inference Endpoints.

“Combining Hugging Face’s embeddings with Elastic’s retrieval relevance tools helps users gain better insights and improve search functionality,” said Jeff Boudier, head of product at Hugging Face. “Hugging Face makes it easy for developers to build their own AI. With this integration, developers get a complete solution to leverage the best open models for semantic search, hosted on Hugging Face multi-cloud GPU infrastructure, to build semantic search experiences in Elasticsearch without worrying about storing or chunking embeddings.”

“Developers are at the heart of our business, and extending more of our GenAI and search primitives to Hugging Face developers deepens our collaboration,” said Matt Riley, global vice president & general manager of search at Elastic. “The integration of our new semantic_text field, simplifies the process of chunking and storing embeddings, so developers can focus on what matters most, building great applications.”

The integration of semantic_text support follows the addition of Hugging Face embeddings models to Elastic’s Open Inference API.

Read the Elastic blog for more information.

About Elastic

Elastic (NYSE: ESTC), the Search AI Company, enables everyone to find the answers they need in real-time using all their data, at scale. Elastic’s solutions for search, observability and security are built on the Elastic Search AI Platform, the development platform used by thousands of companies, including more than 50% of the Fortune 500. Learn more at elastic.co.

Elastic and associated marks are trademarks or registered trademarks of Elastic N.V. and its subsidiaries. All other company and product names may be trademarks of their respective owners.

Contacts

Elastic PR
PR-team@elastic.co

Elastic N.V.

NYSE:ESTC

Release Versions

Contacts

Elastic PR
PR-team@elastic.co

More News From Elastic N.V.

Elastic Announces Multi-Year Partnership with Audi Revolut F1® Team, Expanding Search and Observability Collaboration

SAN FRANCISCO--(BUSINESS WIRE)--Elastic (NYSE: ESTC) and today announced a new multi-year partnership with Audi Revolut F1® Team, building on the team’s existing use of Elastic technology to support its engineering software operations. Elastic joins as Official Race Partner for the 2026 United States Grand Prix in Austin, Texas, and will become its Official Search and Observability Supplier starting from the 2027 season. The agreement brings an established technical collaboration onto a global...

Elastic Introduces Elastic nightshift AI SRE to Investigate with the Full Context Others Can't Afford to Keep

SAN FRANCISCO--(BUSINESS WIRE)--Elastic (NYSE: ESTC) today announced the private preview of Elastic® nightshift AI SRE, an AI agent built into Elastic Observability that investigates like an engineer and shows its work. It continuously detects and investigates issues across the entire stack with full context, including surfacing unknowns that rule-based alerts miss. Elastic nightshift AI SRE tests multiple root-cause hypotheses in parallel against telemetry and other relevant context to deliver...

Elastic Announces AlertZero, a Team of Specialized AI Agents for the Security Operations Lifecycle

SAN FRANCISCO--(BUSINESS WIRE)--Elastic (NYSE: ESTC) today introduced AlertZero, the agentic layer built into Elastic Security, with a team of specialized agents that learn from and adapt to how a security team already works. The agents handle alert triage, investigation, threat hunting, detection tuning and forensic analysis, so each alert gets an evidence-backed answer. This puts analysts on the path to an equivalent of inbox zero: a short queue of recommended actions instead of thousands of...
Back to Newsroom