Skip to content

Conversation

@sdesai345
Copy link
Contributor

Reason for Change:
New conceptual guide to add context and visual diagrams for ./gateway-api-inference-extension.md

@kaito-pr-agent
Copy link

Title

Add conceptual guide for inference traffic routing with GAIE and KAITO


Description

  • Added conceptual guide for inference traffic routing

  • Detailed KAITO integration with Gateway API Inference Extension

  • Explained core concepts and request flow with diagrams

  • Provided advanced use cases and implementation guidance


Changes walkthrough 📝

Relevant files
Documentation
inference-traffic-routing.md
New conceptual guide for inference traffic routing             

website/docs/inference-traffic-routing.md

  • Added new conceptual documentation file
  • Detailed KAITO-GAIE integration architecture
  • Explained core concepts: InferencePool, InferenceModel, Endpoint
    Picker
  • Provided visual diagrams of system architecture and request flow
  • +161/-0 

    Need help?
  • Type /help how to ... in the comments thread for any questions about PR-Agent usage.
  • Check out the documentation for more information.
  • @kaito-pr-agent
    Copy link

    PR Reviewer Guide 🔍

    Here are some key observations to aid the review process:

    ⏱️ Estimated effort to review: 2 🔵🔵⚪⚪⚪
    🧪 No relevant tests
    🔒 No security concerns identified
    ⚡ Recommended focus areas for review

    Broken Link

    The link to ./gateway-api-inference-extension.md might be broken since the target file isn't created in this PR.

    Learn how to enable GAIE and get started with sample deployments on your Kubernetes cluster using [Gateway API Inference Extension with KAITO guide](./gateway-api-inference-extension.md).
    Diagram Inconsistency

    The Endpoint Picker Pod box uses inconsistent hyphen patterns that may break diagram rendering.

    +------------------------+           +--------------------------------+
    |     InferenceModel     | ------->  |  HTTPRoute (Gateway API)       |
    +------------------------+           |  - Route rules                 |
    Formatting Issue

    Extra trailing spaces in the GAIE Controller box may cause misalignment in rendered diagrams.

    | - Model metadata          |  

    @chewong chewong changed the title New conceptual guide for GAIE and KAITO docs: New conceptual guide for GAIE and KAITO Sep 22, 2025
    v | | HelmRelease | |
    +----------------------------+ | +----------------------+ |
    | Model server pod(s) | <------+-----------------------------+
    | - vLLM / Transformers |
    Copy link
    Collaborator

    Choose a reason for hiding this comment

    The reason will be displayed to describe this comment to others. Learn more.

    GAIE doesn't support Hugging Face Transformers since it doesn't provide an OpenAI-compatible API

    +-----------------+
    |
    | HTTP request: POST /v1/chat/completions
    | Headers: model=phi-4-mini-instruct
    Copy link
    Collaborator

    Choose a reason for hiding this comment

    The reason will be displayed to describe this comment to others. Learn more.

    Suggested change
    | Headers: model=phi-4-mini-instruct
    | Body: "model": "phi-4-mini-instruct"

    Copy link
    Collaborator

    Choose a reason for hiding this comment

    The reason will be displayed to describe this comment to others. Learn more.

    You will need to add this page to https://github.com/kaito-project/kaito/blob/main/website/sidebars.js so we can show it in the sidebar

    Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

    Projects

    Status: No status

    Development

    Successfully merging this pull request may close these issues.

    2 participants