Fix hf_quant_config with kv cache type [OMNIML-2918] #557
Conversation
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
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #557 +/- ##
=======================================
Coverage 74.64% 74.64%
=======================================
Files 183 183
Lines 18547 18547
=======================================
Hits 13844 13844
Misses 4703 4703 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
meenchen
reviewed
Nov 14, 2025
| kv_cache_quantization = None | ||
| if get_kv_cache_dtype(self.model) == KV_CACHE_FP8: | ||
| # Only FP8 KV Cache is supported in VLLM for now | ||
| kv_cache_quantization = "FP8" |
Contributor
There was a problem hiding this comment.
Could you also add FP4 KV support? TRT-LLM actually supports FP4 kv now.
meenchen
approved these changes
Nov 14, 2025
jenchen13
force-pushed
the
jennifchen/fix_kv_export
branch
from
478abad to
c2b254c
Compare
jenchen13
changed the title
jenchen13
force-pushed
the
jennifchen/fix_kv_export
branch
from
18852fe to
3ecca22
Compare
Signed-off-by: Jennifer Chen <[email protected]>
jenchen13
force-pushed
the
jennifchen/fix_kv_export
branch
from
3ecca22 to
3cdb810
Compare
Collaborator
|
/ok to test 3cdb810 |
kevalmorabia97
pushed a commit
that referenced
this pull request
Dec 7, 2025
Update hf_quant_config with correct kv cache type for FP8 and NVFP4 --------- Signed-off-by: jenchen13 <[email protected]> Signed-off-by: Jennifer Chen <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
What does this PR do?
Type of change: ? Bug fix
Fix hf_quant_config with correct kv cache type for FP8/NVFP4
Overview: ?
Usage
# Add a code snippet demonstrating how to use thisTesting
will test export with KV cache fp8 enabled
Before your PR is "Ready for review"
Additional Information