Cache (3/5): Cache inference across model providers - #123
Merged
Merged
Conversation
ErlisLushtaku
force-pushed
the
cache-on-118/02-do-inference
branch
from
September 9, 2026 13:01
97138e9 to
965bacf
Compare
ErlisLushtaku
force-pushed
the
cache-on-118/03-providers
branch
from
September 9, 2026 13:01
df52a6d to
e28ee43
Compare
ErlisLushtaku
changed the title
ErlisLushtaku
changed the title
ErlisLushtaku
marked this pull request as ready for review
September 15, 2026 07:40
kargibora
force-pushed
the
cache-on-118/02-do-inference
branch
from
September 15, 2026 12:54
58f0f59 to
969347c
Compare
kargibora
force-pushed
the
cache-on-118/03-providers
branch
from
September 15, 2026 12:54
595f5ce to
1389840
Compare
kargibora
reviewed
Sep 15, 2026
| return engine_kwargs | ||
|
|
||
|
|
||
| _VLLM_ONLY_KWARGS = ( |
Member
There was a problem hiding this comment.
I am not sure with this design. It is very hard-coded. Do we really need to check this?
(Even if this exist or introduced before)
Collaborator
Author
There was a problem hiding this comment.
I agree it seems not very general, but I kept this because shared engine_kwargs can include vLLM-only settings for hosted providers, which would otherwise receive them as request parameters. If you have any different idea how to address this, let me know.
Member
There was a problem hiding this comment.
Yes I agree. Perhaps this is a matter of another PR.
…to cache-on-118/03-providers # Conflicts: # judgearena/inference.py # tests/test_inference_cache.py
…o cache-on-118/02-do-inference
…to cache-on-118/03-providers
kargibora
approved these changes
Sep 21, 2026
ErlisLushtaku
changed the base branch from
cache-on-118/02-do-inference
to
main
September 22, 2026 22:30
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.
Description
Extends cache descriptors and input hashing across the providers exposed by
make_model.extra_body, and output-affecting model settings.tensor_parallel_size,gpu_memory_utilization,enforce_eager).This is stacked on #122.