Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[ROCm][AMD][Model] llama 3.2 support upstreaming #12421

Merged
merged 6 commits into from
Jan 31, 2025
Merged

[ROCm][AMD][Model] llama 3.2 support upstreaming #12421

merged 6 commits into from
Jan 31, 2025

Conversation

maleksan85
Copy link
Contributor

PR to propagate multimodal llama3.2 support into upstream for rocm arch

Aleksandr Malyshev added 2 commits January 24, 2025 13:37
initial commit with rocm fa update
d5b47dc 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
merge with main
1f34f4e 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
@github-actions GitHub Actions
Copy link

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

  • Add ready label to the PR
  • Enable auto-merge.

🚀

linters
d8aa3de 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
@maleksan85 maleksan85 changed the title ROCM llama 3.2 support upstreaming [ROCm][AMD][Model] llama 3.2 support upstreaming Jan 24, 2025
heheda12345
heheda12345 suggested changes Jan 25, 2025
View reviewed changes
Copy link
Collaborator

@heheda12345 heheda12345 left a comment

Choose a reason for hiding this comment

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

Thanks for contributing. Added some comments for this PR.

vllm/model_executor/models/mllama.py Outdated
i,
)
elif self.attn.backend in (_Backend.XFORMERS, _Backend.TORCH_SDPA):
if current_platform.is_rocm():
Copy link
Collaborator

Choose a reason for hiding this comment

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

Can you merge this code path with if self.attn.backend in (_Backend.FLASH_ATTN, _Backend.FLASH_ATTN_VLLM_V1):?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

normally I would not change code as purpose of this PR it is to match rocm repo and upstream in order not to bring more problems. But code seems identical in two branches and no reason to keep it both. Thank you for it!

replying to comments
f1f62e6 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
@hongxiayang hongxiayang added the rocm Related to AMD ROCm label Jan 27, 2025
@DarkLight1337
Copy link
Member

LGTM, can you merge from main to fix the CI failures?

Aleksandr Malyshev added 2 commits January 28, 2025 19:12
Merge branch 'upstream/main' into rocm/maleksan_llama32_upstreaming
ccc5285
linter
e096bc8 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
@maleksan85
Copy link
Contributor Author

@DarkLight1337 do you have any other concerns about this change? would be nice to land it asap to prevent merge conflicts. Thanks

@DarkLight1337 DarkLight1337 enabled auto-merge (squash) January 29, 2025 17:28
@github-actions github-actions bot added the ready ONLY add when PR is ready to merge/full CI is needed label Jan 29, 2025
@DarkLight1337
Copy link
Member

I have started the merge process.

DarkLight1337
DarkLight1337 approved these changes Jan 31, 2025
View reviewed changes
Copy link
Member

@DarkLight1337 DarkLight1337 left a comment

Choose a reason for hiding this comment

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

LGTM, thanks for your patience!

@DarkLight1337 DarkLight1337 merged commit a1fc18c into vllm-project:main Jan 31, 2025
65 checks passed
Isotr0py pushed a commit to Isotr0py/vllm that referenced this pull request Feb 2, 2025
@maleksan85 @Isotr0py
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
6eaff0b 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
youngkent pushed a commit to youngkent/vllm that referenced this pull request Feb 3, 2025
@maleksan85 @youngkent
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
661ab75 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
srikanthsrnvs pushed a commit to srikanthsrnvs/vllm that referenced this pull request Feb 3, 2025
@maleksan85 @srikanthsrnvs
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
9813962 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
Signed-off-by: Srikanth Srinivas <srikanth@astrum.ai>
NickLucche pushed a commit to NickLucche/vllm that referenced this pull request Feb 7, 2025
@maleksan85 @NickLucche
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
d2615b3 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
ShangmingCai pushed a commit to ShangmingCai/vllm that referenced this pull request Feb 10, 2025
@maleksan85 @ShangmingCai
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
1ea5aef 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
GWS0428 pushed a commit to GWS0428/VARserve that referenced this pull request Feb 12, 2025
@maleksan85 @GWS0428
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
3027c14 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
panf2333 pushed a commit to yottalabsai/vllm that referenced this pull request Feb 18, 2025
@maleksan85 @panf2333
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
18b4fb6 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
kerthcet pushed a commit to kerthcet/vllm that referenced this pull request Feb 21, 2025
@maleksan85 @kerthcet
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
321fef9 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
lk-chen pushed a commit to lk-chen/vllm that referenced this pull request Mar 5, 2025
@maleksan85 @lk-chen
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
da05cc3 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
Signed-off-by: Linkun Chen <github@lkchen.net>
Said-Akbar pushed a commit to Said-Akbar/vllm-rocm that referenced this pull request Mar 7, 2025
@maleksan85 @Said-Akbar
[ROCm][AMD][Model] llama 3.2 support upstreaming (vllm-project#12421)
fa4b067 
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
Signed-off-by: saeediy <saidakbarp@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@heheda12345 heheda12345 heheda12345 requested changes

@DarkLight1337 DarkLight1337 DarkLight1337 approved these changes

Assignees
No one assigned
Labels
ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@maleksan85 @DarkLight1337 @heheda12345 @hongxiayang