Hi all 👋 I’d like to request a review for the following PR: [FLINK-38857][Model] Add comprehensive docs for Triton inference model PR: https://github.com/apache/flink/pull/27490 What this PR does This is a documentation-only PR that introduces a comprehensive guide for using Triton inference models in Flink SQL, including: New documentation page under

docs/connectors/models/triton

End-to-end SQL examples for model definition and inference Advanced configuration (auth, batching, compression, priority, headers) Stateful / sequence model examples Array-type inference examples Performance considerations & troubleshooting Chinese documentation aligned with EN version • Minor updates to

CREATE MODEL

/

ML_PREDICT

docs to reference Triton No runtime or API changes are introduced. Why this matters Triton is a widely adopted inference serving system. This documentation aims to: • Lower the barrier for integrating real-time inference into Flink SQL pipelines • Clarify configuration best practices • Improve overall model inference documentation consistency CR feedback is highly appreciated, especially regarding: • Documentation structure & clarity • Consistency with existing model docs (e.g., OpenAI) • Terminology and SQL examples • Translation accuracy (EN/ZH consistency) Thanks in advance 🙏