ci: Update sm12X minimum cuda capability to 12.9 in aot.py by bkryu · Pull Request #2188 · flashinfer-ai/flashinfer

bkryu · 2025-12-08T22:10:19Z

📌 Description

CUDA Release notes indicate full sm120 support on CUDA 12.8 and sm121 on 12.9. Changing the version noted in aot.py.

The change alters the coverage in flashinfer-jit-cache -- today some sm120 kernels are not compiled & shipped in cuda 12.9, but are included in 13.0. The change here will insure kernels are included.

🔍 Related Issues

🚀 Pull Request Checklist

Thank you for contributing to FlashInfer! Before we review your pull request, please make sure the following items are complete.

✅ Pre-commit Checks

I have installed pre-commit by running pip install pre-commit (or used your preferred method).
I have installed the hooks with pre-commit install.
I have run the hooks manually with pre-commit run --all-files and fixed any reported issues.

If you are unsure about how to set up pre-commit, see the pre-commit documentation.

🧪 Tests

Tests have been added or updated as needed.
All tests are passing (unittest, etc.).

Reviewer Notes

Summary by CodeRabbit

Bug Fixes
- Adjusted GPU capability detection and CUDA version requirements:
  - SM110 now requires CUDA 13.0 (raised from 12.9).
  - SM120 support expanded to CUDA 12.8 (lowered from 13.0).
  - SM121 support updated to CUDA 12.9 (from 13.0).
Impact
- Broader compatibility for SM120/SM121 on slightly older CUDA; SM110 now requires a newer CUDA runtime.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

coderabbitai · 2025-12-08T22:10:29Z

Walkthrough

Adjusts CUDA SM capability detection in flashinfer/aot.py by updating the version thresholds for SM110, SM120, and SM121 mappings (tuning which CUDA runtimes are considered compatible). No other logic or control-flow changes present.

Changes

Cohort / File(s)	Summary
SM capability detection `flashinfer/aot.py`	Update CUDA version thresholds for SM mappings: `sm110` threshold changed from 12.9 → 13.0; `sm120` threshold changed from 13.0 → 12.8; `sm121` threshold changed from 13.0 → 12.9. No other code paths modified.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Verify the new numeric thresholds match intended CUDA release/version semantics.
Check any tests or runtime checks that assume previous thresholds.
Confirm downstream logic (feature gating or selection) still behaves correctly with updated mappings.

Possibly related PRs

[NVIDIA] Thor & Spark Support #2028 — Modifies CUDA compute-capability/version handling and SM mappings; closely related to these threshold adjustments.

Suggested reviewers

aleozlx
yzh119
cyx-6
wenscarl
nvmbreughe

Poem

🐰 I hopped through versions, nibbling fine,
Swapped some numbers — 12.9 to 13.0, align!
SM120 and 121 now dance in line,
A tiny tweak, a tidy sign,
Bright kernels hop and sing with brine. 🥕

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly describes the main change: updating sm12X minimum CUDA capability to 12.9 in aot.py, which matches the actual changes in the PR.
Description check	✅ Passed	The description includes a detailed explanation of the changes referencing CUDA release notes and explaining the impact on flashinfer-jit-cache coverage, with pre-commit checklist completed and tests appropriately not marked.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

✨ Finishing touches

📝 Generate docstrings

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment

📜 Recent review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 35c0347 and 1b95a22.

📒 Files selected for processing (1)

flashinfer/aot.py (1 hunks)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

GitHub Check: Deploy Docs

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

gemini-code-assist · 2025-12-08T22:10:30Z

Summary of Changes

Hello @bkryu, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refines the configuration for CUDA compute capabilities within the aot.py file. Specifically, it adjusts the minimum required CUDA version for sm120 and sm121 architectures to 12.9, potentially enhancing compatibility or correcting a previous version specification for these compute units.

Highlights

CUDA Capability Update: The minimum CUDA capability for sm120 and sm121 architectures has been updated from 13.0 to 12.9 in aot.py.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request updates the minimum required CUDA version for sm120 and sm121 architectures from 13.0 to 12.9 in flashinfer/aot.py. This change appears to be a correction to enable compilation for these architectures on CUDA 12.9, which is consistent with the versions for other newer SM architectures like sm100f, sm103, and sm110. The change is straightforward and I don't see any issues with it.

bkryu · 2025-12-08T22:12:17Z

/bot run

flashinfer-bot · 2025-12-08T22:13:32Z

GitLab MR !183 has been created, and the CI pipeline #39835910 is currently running. I'll report back once the pipeline job completes.

bkryu · 2025-12-08T22:24:27Z

/bot run

flashinfer-bot · 2025-12-08T22:24:53Z

GitLab MR !183 has been updated with latest changes, and the CI pipeline #39836651 is currently running. I'll report back once the pipeline job completes.

yzh119

LGTM!

…r-ai#2188)  ## 📌 Description [CUDA Release notes](https://docs.nvidia.com/cuda/parallel-thread-execution/#release-notes) indicate full sm120 support on CUDA 12.8 and sm121 on 12.9. Changing the version noted in `aot.py`. The change alters the coverage in `flashinfer-jit-cache` -- today some sm120 kernels are not compiled & shipped in cuda 12.9, but are included in 13.0. The change here will insure kernels are included.  ## 🔍 Related Issues  ## 🚀 Pull Request Checklist Thank you for contributing to FlashInfer! Before we review your pull request, please make sure the following items are complete. ### ✅ Pre-commit Checks - [x] I have installed `pre-commit` by running `pip install pre-commit` (or used your preferred method). - [x] I have installed the hooks with `pre-commit install`. - [x] I have run the hooks manually with `pre-commit run --all-files` and fixed any reported issues. > If you are unsure about how to set up `pre-commit`, see [the pre-commit documentation](https://pre-commit.com/). ## 🧪 Tests - [ ] Tests have been added or updated as needed. - [ ] All tests are passing (`unittest`, etc.). ## Reviewer Notes   ## Summary by CodeRabbit * **Bug Fixes** * Adjusted GPU capability detection and CUDA version requirements: * SM110 now requires CUDA 13.0 (raised from 12.9). * SM120 support expanded to CUDA 12.8 (lowered from 13.0). * SM121 support updated to CUDA 12.9 (from 13.0). * **Impact** * Broader compatibility for SM120/SM121 on slightly older CUDA; SM110 now requires a newer CUDA runtime. <sub>✏️ Tip: You can customize this high-level summary in your review settings.</sub>

Update sm120 and 121 with min cuda capability 12.9

35c0347

bkryu requested review from aleozlx, cyx-6, nvmbreughe, wenscarl and yzh119 as code owners December 8, 2025 22:10

gemini-code-assist Bot reviewed Dec 8, 2025

View reviewed changes

Fix 110 and 120

1b95a22

yzh119 approved these changes Dec 9, 2025

View reviewed changes

yzh119 merged commit 6bb01d1 into flashinfer-ai:main Dec 9, 2025
4 checks passed

bkryu deleted the aot_sm12x_capability branch December 9, 2025 22:02

coderabbitai Bot mentioned this pull request Feb 17, 2026

feat: add is_sm12x_supported() helper for SM12x family detection #2574

Merged

coderabbitai Bot mentioned this pull request Mar 3, 2026

Enable sm120f compilation #2650

Merged

5 tasks

This was referenced Mar 16, 2026

[Spark unit test debugging] Fix for tests/gemm/test_groupwise_scaled_gemm_fp8.py #2751

Merged

[Spark bug] Fix arch 12.1 -> "sm120a" flag for Spark, CUDA 12.9 #2839

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

ci: Update sm12X minimum cuda capability to 12.9 in aot.py#2188

ci: Update sm12X minimum cuda capability to 12.9 in aot.py#2188
yzh119 merged 2 commits intoflashinfer-ai:mainfrom
bkryu:aot_sm12x_capability

bkryu commented Dec 8, 2025 •

edited

Loading

Uh oh!

coderabbitai Bot commented Dec 8, 2025 •

edited

Loading

Uh oh!

gemini-code-assist Bot commented Dec 8, 2025

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

bkryu commented Dec 8, 2025

Uh oh!

flashinfer-bot commented Dec 8, 2025

Uh oh!

bkryu commented Dec 8, 2025

Uh oh!

flashinfer-bot commented Dec 8, 2025

Uh oh!

yzh119 left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

bkryu commented Dec 8, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📌 Description

🔍 Related Issues

🚀 Pull Request Checklist

✅ Pre-commit Checks

🧪 Tests

Reviewer Notes

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Dec 8, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Estimated code review effort

Possibly related PRs

Suggested reviewers

Poem

Pre-merge checks and finishing touches

Uh oh!

gemini-code-assist Bot commented Dec 8, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

bkryu commented Dec 8, 2025

Uh oh!

flashinfer-bot commented Dec 8, 2025

Uh oh!

bkryu commented Dec 8, 2025

Uh oh!

flashinfer-bot commented Dec 8, 2025

Uh oh!

yzh119 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

bkryu commented Dec 8, 2025 •

edited

Loading

coderabbitai Bot commented Dec 8, 2025 •

edited

Loading