chore: bumps vllm to v0.24.0#1026
Merged
Merged
Conversation
gkumbhat
requested review from
joerunde,
prashantgupta24,
rafvasq,
sducouedic,
tdoublep and
yannicks1
as code owners
|
👋 Hi! Thank you for contributing. We also recommend installing prek and configuring it to check your code before every local commit. |
joerunde
reviewed
Jul 6, 2026
| # initializing. Importing vllm.tokenizers.registry here would cause a circular | ||
| # import, so the call is deferred to SpyrePlatform.check_and_update_config. | ||
| # Workers use model execution paths, not the tokenizer registry, so they do | ||
| # not need this patch applied at module-level. |
Collaborator
There was a problem hiding this comment.
@gkumbhat this entire comment block should be moved to check_and_update_config then
Also- I'm really dubious about this working since the previous comment explains that this had to be here in order to be picked up by the spawned worker processes. If those don't run check_and_update_config because the config is simply serialized and sent to them, then the patch won't apply
Defers tokenizer registry patch from module-level to check_and_update_config to fix a circular import introduced in vLLM 0.24.0, where torch_utils.py now imports is_pin_memory_available at module-level (triggering platform resolution before torch_utils finishes initializing). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Collaborator
|
bot:test |
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Collaborator
Author
|
bot:test |
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Both are now available unconditionally in vLLM 0.24.0 (the new minimum). Also updates vLLM:lowest CI entry to v0.24.0 and drops all pre-0.24.0 intermediate matrix entries. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com>
github-actions
Bot
added
the
ready
gkumbhat
added a commit
that referenced
this pull request
Jul 7, 2026
Defers tokenizer registry patch from module-level to check_and_update_config to fix a circular import introduced in vLLM 0.24.0, where torch_utils.py now imports is_pin_memory_available at module-level (triggering platform resolution before torch_utils finishes initializing). <!-- markdownlint-disable --> Enable `0.24.0` vLLM 1. vLLM upstream PR vllm-project/vllm#42656 seems to have slightly changed how blocks gets prepended, which seems to fail the tests. 2. vLLM upstream PR vllm-project/vllm#40299 seems to have fixed the mistral model file parsing issue, thus with move to `0.24.0` we can remove the patch we added for that entirely. <!-- Link related issues, e.g., `Fixes #` or `Relates to #456` --> <!-- Describe how you tested your changes. Include commands or steps to reproduce. --> - [ ] I have read the [contributing guidelines](https://docs.vllm.ai/projects/spyre/en/latest/contributing) - [ ] My code follows the project's code style (run `bash format.sh`) - [ ] I have added tests for my changes (if applicable) - [ ] I have updated the documentation (if applicable) - [ ] My commits include a `Signed-off-by:` line (DCO compliance) --------- Signed-off-by: gkumbhat <kumbhat.gaurav@gmail.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Defers tokenizer registry patch from module-level to check_and_update_config to fix a circular import introduced in vLLM 0.24.0, where torch_utils.py now imports is_pin_memory_available at module-level (triggering platform resolution before torch_utils finishes initializing).
Description
Enable
0.24.0vLLM0.24.0we can remove the patch we added for that entirely.Related Issues
Test Plan
Checklist
bash format.sh)Signed-off-by:line (DCO compliance)