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.

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

tgi: revert xeon version to 2.2.0 #328

Merged
merged 2 commits into from
Aug 21, 2024
Merged

Conversation

lianhao
Copy link
Collaborator

@lianhao lianhao commented Aug 21, 2024

Description

Due to bug opea-project/GenAIExamples#636, revert tgi xeon version to 2.2.0.

Issues

opea-project/GenAIExamples#636

Type of change

List the type of change like below. Please delete options that are not relevant.

  • Bug fix (non-breaking change which fixes an issue)

Dependencies

n/a.

Tests

Describe the tests that you ran to verify your changes.

Due to bug opea-project/GenAIExamples#636, revert tgi xeon version to
2.2.0.

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Copy link
Contributor

@daisy-ycguo daisy-ycguo left a comment

Choose a reason for hiding this comment

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

lgtm

@lianhao lianhao added this to the v0.9 milestone Aug 21, 2024
@lianhao
Copy link
Collaborator Author

lianhao commented Aug 21, 2024

The reason why gmc xeon test fails is because the model used by gmc chatqna CI test bigscience/bloom-560m doesn't work with tgi 2.2.0 anymore. We use bigscience/bloom-560m in GMC xeon CI because of bug #258.

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
@KfreeZ
Copy link
Collaborator

KfreeZ commented Aug 21, 2024

note: TGI 2.2.0 cannot work with bigscience/bloom-560m

Copy link
Collaborator

@KfreeZ KfreeZ 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 fixing this!!!

@KfreeZ KfreeZ merged commit 076e81e into opea-project:main Aug 21, 2024
14 checks passed
@lianhao lianhao deleted the tgi_2.2.0 branch August 21, 2024 07:41
@eero-t
Copy link
Contributor

eero-t commented Aug 21, 2024

The reason why gmc xeon test fails is because the model used by gmc chatqna CI test bigscience/bloom-560m doesn't work with tgi 2.2.0 anymore. We use bigscience/bloom-560m in GMC xeon CI because of bug #258.

Whereas 2.0.x & 2.1.x versions regressed TGI Prometheus support so one does not get any token metrics out of them (although they worked fine with TGI 1.x). That was fixed only in 2.2.0: huggingface/text-generation-inference#2184

chensuyue pushed a commit that referenced this pull request Aug 21, 2024
* tgi: revert xeon version to 2.2.0

Due to bug opea-project/GenAIExamples#636, revert tgi xeon version to
2.2.0.

* Retriever: Fix missing HF_TOKEN issue of v0.9 retriever-redis image

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
(cherry picked from commit 076e81e)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants