← All updates Databricks Update One-Pager Navy version
databricks Platform · Updated Aug 7, 2025

Use OpenAI GPT OSS models for batch inference and structured outputs

The Databricks hosted foundation models, OpenAI GPT OSS 120B and GPT OSS 20B are now optimized for AI Functions, which means you can perform batch inference using these models and AI Functions like…

Update Open docs ↗release note ↗
What it does

The Databricks hosted foundation models, OpenAI GPT OSS 120B and GPT OSS 20B are now optimized for AI Functions, which means you can perform batch inference using these models and AI Functions like ai query(). Inference workloads that use these models also now support specifying structured outputs. Applies to: Databricks SQL Databricks Runtime Invokes an existing Databricks Model Serving endpoint and parses and returns its response.

Why it matters

Reduces operational overhead and unlocks new capabilities for data teams on Databricks.

Key capabilities
DS fixes grammar issues in
DS fixes grammar issues in a batch. > SELECT EXCEPT text, correctgrammar(text) AS text FROM articles;
Recommended use cases
SQL SELECT textrequest = named struct( 'timestamp'timestamp'sender'
Architecture
Databricks Platform → Platform
In short

The Databricks hosted foundation models, OpenAI GPT OSS 120B and GPT OSS 20B are now optimized for AI Functions, which means you can perform batch inference using these models and AI Functions like ai query(). Inference workloads that use these models also…

Read the full documentation ↗
Key takeaway
The Databricks hosted foundation models, OpenAI GPT OSS 120B and GPT OSS 20B are now optimized for AI Functions, which means you can perform batch inference using these models and AI Functions like…
2025-08-07-use-openai-gpt-oss-models-for-batch-inference-and-structured-outputs Generated by the Databricks Updates Agent · browse all updates