Skip to content

docs(FR-2603): apply 26.4 model serving doc feedback — service launcher fields, vLLM/SGLang runtime params#6804

Open
agatha197 wants to merge 1 commit into04-17-docs_fr-2571_address_26.4_docs_feedback_remove_legacy_model_serving_sections_and_apply_ko_wording_fixesfrom
docs/FR-2603-apply-26.4-model-serving-doc-feedback
Open

docs(FR-2603): apply 26.4 model serving doc feedback — service launcher fields, vLLM/SGLang runtime params#6804
agatha197 wants to merge 1 commit into04-17-docs_fr-2571_address_26.4_docs_feedback_remove_legacy_model_serving_sections_and_apply_ko_wording_fixesfrom
docs/FR-2603-apply-26.4-model-serving-doc-feedback

Conversation

@agatha197
Copy link
Copy Markdown
Contributor

@agatha197 agatha197 commented Apr 17, 2026

Resolves #6803 (FR-2603)

Summary

Addresses additional Sujin feedback on the 26.4 model serving docs, stacked on top of #6778.

  • Service Launcher fields: Refined descriptions for Service Name, Open To Public, Model Storage Folder to Mount, Inference Runtime Variant, and Environment / Version across en/ja/ko/th.
  • Runtime parameters: Split into vLLM and SGLang subsections reflecting the new tab-based UI (Model Loading / Resource Memory / Serving Performance / Multimodal / Tool Reasoning). Documented Model Loading tab fields (Model, DType, Quantization, Max Model Length / Context Length, Served Model Name, Trust Remote Code).
  • Legacy section removal: Removed the <details> block about model service configuration in version 23.03 and earlier. Moved the Custom runtime workflow (model / service definition files) behind a collapsible block below the 26.4 quick-deploy flow so the default path is visible first.
  • Command mode section: Moved the command-mode screenshot before the field list and consolidated the field bullets (start command, model mount, port auto-detect, health check URL auto-detect, initial delay, max retries).
  • Port / Health URL / Mount auto-detection: Clarified that these fields are auto-detected from the start command.
  • Scale Out/In sentence removal: Removed the sentence referencing the explicit Scale Out / Scale In selector, since the step-size column now infers direction automatically.
  • Drawer error alert rendering fix: Changed *No compatible presets available. This model cannot be deployed.* to **...** so it renders as bold (the original italic was not rendering reliably next to CJK characters).
  • Screenshots: Refreshed service_launcher1.png and service_launcher_command_mode.png for ko. Added new service_launcher_runtime_params_vllm.png and service_launcher_runtime_params_sglang.png for ko.

Test plan

  • Render the ko docs locally and verify the drawer error alert in the Model Store section renders as bold (no stray * characters).
  • Verify the Service Launcher field list and the command-mode subsection show the updated labels and screenshots.
  • Verify the vLLM and SGLang runtime parameters subsections render with the new per-tab field descriptions and the new screenshots.
  • Regenerate the ko PDF (pnpm run pdf:ko) and confirm pages 121–122 show the updated vLLM / SGLang runtime parameter images and that 그림 15.6 shows the updated service launcher view.

Copilot AI review requested due to automatic review settings April 17, 2026 14:03
@github-actions github-actions Bot added the size:XL 500~ LoC label Apr 17, 2026
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

Updates the Backend.AI WebUI user manual (Model Serving) to reflect the 26.4 Service Launcher UI changes, including revised field descriptions and the new tab-based runtime parameters for vLLM and SGLang, with corresponding screenshot updates.

Changes:

  • Reworked the “steps/workflow” section to emphasize 26.4 quick deploy + Service Launcher, and moved Custom runtime file-based workflow into a collapsible “Advanced” block.
  • Updated Service Launcher field descriptions and the command-mode field list (auto-detected Port/Health URL/Mount, plus retry/delay fields).
  • Split runtime parameters into vLLM vs SGLang subsections and added/updated screenshots (notably in ko).

Reviewed changes

Copilot reviewed 4 out of 8 changed files in this pull request and generated 13 comments.

Show a summary per file
File Description
packages/backend.ai-webui-docs/src/en/model_serving/model_serving.md Reorganized workflow/advanced section; updated launcher fields and runtime parameter docs (vLLM/SGLang).
packages/backend.ai-webui-docs/src/ja/model_serving/model_serving.md Japanese translation updates aligned to the new launcher/runtime-params UI structure.
packages/backend.ai-webui-docs/src/ko/model_serving/model_serving.md Korean updates for launcher fields/runtime params; also adjusts Model Store error emphasis.
packages/backend.ai-webui-docs/src/th/model_serving/model_serving.md Thai translation updates aligned to the new launcher/runtime-params UI structure.
packages/backend.ai-webui-docs/src/ko/images/service_launcher_command_mode.png Updated Service Launcher command-mode screenshot asset.

Comment thread packages/backend.ai-webui-docs/src/ko/model_serving/model_serving.md Outdated
Comment thread packages/backend.ai-webui-docs/src/ja/model_serving/model_serving.md Outdated
Comment thread packages/backend.ai-webui-docs/src/ja/model_serving/model_serving.md Outdated
Comment thread packages/backend.ai-webui-docs/src/th/model_serving/model_serving.md Outdated
Comment thread packages/backend.ai-webui-docs/src/th/model_serving/model_serving.md Outdated
Copy link
Copy Markdown
Contributor Author

agatha197 commented Apr 21, 2026

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more


How to use the Graphite Merge Queue

Add either label to this PR to merge it via the merge queue:

  • flow:merge-queue - adds this PR to the back of the merge queue
  • flow:hotfix - for urgent changes, fast-track this PR to the front of the merge queue

You must have a Graphite account in order to use the merge queue. Sign up using this link.

An organization admin has required the Graphite Merge Queue in this repository.

Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.

This stack of pull requests is managed by Graphite. Learn more about stacking.

@agatha197 agatha197 force-pushed the 04-17-docs_fr-2571_address_26.4_docs_feedback_remove_legacy_model_serving_sections_and_apply_ko_wording_fixes branch from 62feb55 to a914365 Compare April 21, 2026 11:07
@agatha197 agatha197 force-pushed the docs/FR-2603-apply-26.4-model-serving-doc-feedback branch from 467ad1b to 3d3f3cc Compare April 21, 2026 11:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

size:XL 500~ LoC

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants