[None][doc] Add Qwen3.8 deployment guide and configs - #17594
Conversation
|
/bot run --disable-fail-fast |
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Enterprise Run ID: 📒 Files selected for processing (4)
🚧 Files skipped from review as they are similar to previous changes (4)
WalkthroughAdded Qwen3.8 curated serving configurations, repository-relative command handling, and a combined Qwen3.8/Qwen3.5 deployment guide. Updated model support, MTP support, documentation navigation, and configuration metadata. ChangesQwen3.8 deployment support
Estimated code review effort: 3 (Moderate) | ~25 minutes Mergeability Score: 🟡 Moderate · up to The deployment guide and associated configuration document conflicting MTP3 concurrency settings, which could lead users to run misleading benchmarks and make incorrect capacity or performance decisions. Merge should wait for the inconsistency to be corrected or explicitly accepted by the owner. Sequence Diagram(s)sequenceDiagram
participant ConfigTable as generate_config_table.py
participant Selector as config_selector.js
participant Server as trtllm-serve
ConfigTable->>Selector: Generate serving command
Selector->>Selector: Add repository directory when required
Selector->>Server: Run with selected configuration
Suggested reviewers: 🚥 Pre-merge checks | ✅ 3 | ❌ 2❌ Failed checks (2 warnings)
✅ Passed checks (3 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In
`@docs/source/deployment-guide/deployment-guide-for-qwen3.8-qwen3.5-on-trtllm.md`:
- Line 55: Resolve the inconsistency between the MTP3 profile’s max_batch_size,
the concurrency value in the “High throughput” table row, and the attention-DP
scaling rule near the sweep-limit definition. Align these values so the
documented concurrency does not exceed max_batch_size * num_gpus, updating the
performance value, profile setting, or scaling rule as appropriate while
preserving the intended MTP3 benchmark configuration.
In `@scripts/generate_config_table.py`:
- Around line 162-169: Annotate the node parameter of the nested walk function
with object while preserving its existing boolean return annotation and
traversal behavior.
🪄 Autofix
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 9038d75b-d7a3-4934-8c80-dca558953ffe
📒 Files selected for processing (15)
docs/source/_static/config_db.jsondocs/source/_static/config_selector.jsdocs/source/deployment-guide/deployment-guide-for-qwen3.5-on-trtllm.mddocs/source/deployment-guide/deployment-guide-for-qwen3.8-qwen3.5-on-trtllm.mddocs/source/deployment-guide/index.rstdocs/source/features/speculative-decoding.mddocs/source/models/supported-models.mdexamples/configs/curated/eplb/qwen3.8-ep32-static544-mtp3.yamlexamples/configs/curated/eplb/qwen3.8-ep32-static544.yamlexamples/configs/curated/lookup.yamlexamples/configs/curated/qwen3.8-high-throughput-mtp3.yamlexamples/configs/curated/qwen3.8-high-throughput.yamlexamples/configs/curated/qwen3.8-low-latency-mtp3.yamlexamples/configs/curated/qwen3.8-low-latency.yamlscripts/generate_config_table.py
💤 Files with no reviewable changes (1)
- docs/source/deployment-guide/deployment-guide-for-qwen3.5-on-trtllm.md
|
PR_Github #65763 [ run ] triggered by Bot. Commit: |
|
PR_Github #65763 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #65832 [ run ] triggered by Bot. Commit: |
|
PR_Github #65832 [ run ] completed with state
|
6278691 to
e9752cf
Compare
Signed-off-by: Wanli Jiang <35160485+Wanli-Jiang@users.noreply.github.com>
e9752cf to
fad750a
Compare
Wanli-Jiang
changed the title
|
/bot run --disable-fail-fast |
|
PR_Github #65861 [ run ] triggered by Bot. Commit: |
|
PR_Github #65861 [ run ] completed with state
|
|
/bot skip --comment "skip since the failures are not related with code changes" |
|
PR_Github #65925 [ skip ] triggered by Bot. Commit: |
|
PR_Github #65925 [ skip ] completed with state |
3 participants
Dev Engineer Review
QA Engineer Review
No test changes.
Description
Test Coverage
PR Checklist
Please review the following before submitting your PR:
PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
If PR introduces API changes, an appropriate PR label is added - either
api-compatibleorapi-breaking. Forapi-breaking, includeBREAKINGin the PR title.Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
Update tava architecture diagram if there is a significant design change in PR.
The reviewers assigned automatically/manually are appropriate for the PR.
Please check this after reviewing the above items as appropriate for this PR.
GitHub Bot Help
To see a list of available CI bot commands, please comment
/bot help.