Skip to content

Pull requests: vllm-project/compressed-tensors

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

relax setuptools_scm version requirement
#343 opened Jun 6, 2025 by envolution Loading… updated Jun 6, 2025
Optimize sparse 2:4 compression performance
#358 opened Jun 16, 2025 by rahul-tuli Draft updated Jun 16, 2025
8 tasks done
[KV Cache] support kv cache int8 per channel quant
#398 opened Jul 19, 2025 by Eviannn Loading… updated Aug 7, 2025
[FP4] Update to make compression handling more generic for fp4
#448 opened Sep 8, 2025 by dsikka Draft updated Sep 8, 2025
[Observer Refactor] Use static defaults
#489 opened Oct 13, 2025 by kylesayrs Draft updated Oct 15, 2025
[Attention] Support FP4 attention quantization
#491 opened Oct 14, 2025 by kylesayrs Loading… updated Oct 23, 2025
early and better error for divisibility issues
#510 opened Nov 6, 2025 by HDCharles Draft updated Nov 13, 2025
support wInt4aFp8 for moe
#518 opened Nov 12, 2025 by Wangzheee Loading… updated Nov 14, 2025
[Bugfix] Forward quantize better wrapping
#521 opened Nov 18, 2025 by kylesayrs Loading… updated Nov 18, 2025
[Utils] Add return_unmatched argument to match_modules_set
#522 opened Nov 19, 2025 by kylesayrs Loading… updated Nov 19, 2025
[Quantization] Guard against Nan/Inf scales
#523 opened Nov 19, 2025 by kylesayrs Draft updated Nov 19, 2025
fix qparams decompression bug Something isn't working
#514 opened Nov 10, 2025 by shanjiaz Loading… updated Dec 12, 2025
ProTip! Adding no:label will show everything without a label.