Skip to content

Pull requests: vllm-project/llm-compressor

New pull request New
56 Open 1,356 Closed
56 Open 1,356 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Tests][LM-EVAL] Remove / update configs
#2692 opened May 7, 2026 by dsikka Collaborator • Draft
4
[Tests][E2E] Remove / update configs
#2691 opened May 7, 2026 by dsikka Collaborator • Draft
5
Update example autoround README.md autoround For any PR / issue related to autoround support documentation Improvements or additions to documentation enhancement New feature or request fp8 For any issue / PR related to FP8 support nvfp4 For any PR / issue related to NVFP4 support two-reviews When a PR requires two reviews w4a16
#2687 opened May 6, 2026 by changwangss Contributor • Draft
6
[Tracing] Support tracing cache ready When a PR is ready for review Refactor Code cleanup and/or improvements to existing features tracing Issues related to model tracing two-reviews When a PR requires two reviews
#2686 opened May 5, 2026 by kylesayrs Collaborator Loading…
12
70 fix ddp cu dist Work pertaining to distributed work quality-failed ready When a PR is ready for review
#2684 opened May 4, 2026 by HDCharles Collaborator Loading…
18
fix: add enable_thinking flag and reasoning dataset example for Qwen3-Next AWQ awq For any issue / PR related to AWQ support bug Something isn't working documentation Improvements or additions to documentation enhancement New feature or request qwen For any PR / issue related to Qwen support two-reviews When a PR requires two reviews w4a16
#2681 opened May 2, 2026 by jayakumarpujar Contributor Loading…
3 tasks
1
16
Update observer and modifier docs for refactored observer API documentation Improvements or additions to documentation
#2671 opened Apr 30, 2026 by HDCharles Collaborator Loading…
2 tasks
5
feat: concurrent KLD evaluation without enforce_eager (closes #2667, refs #2646) enhancement New feature or request ready When a PR is ready for review two-reviews When a PR requires two reviews
#2668 opened Apr 29, 2026 by jayakumarpujar Contributor Loading…
@HDCharles
12
[Refactor] Consolidate Intermediate Offloading awq For any issue / PR related to AWQ support gptq For any PR / issue related to GPTQ support needs-rebase Refactor Code cleanup and/or improvements to existing features two-reviews When a PR requires two reviews
#2664 opened Apr 28, 2026 by menogrey Contributor Loading…
21
Feat/issue 2646 enhancement New feature or request ready When a PR is ready for review two-reviews When a PR requires two reviews
#2663 opened Apr 28, 2026 by rpathade Loading…
@HDCharles
30
[Examples] Kimi K2.6 enhancement New feature or request fp8 For any issue / PR related to FP8 support model_free_ptq For any PR/issue related to the `model_free_ptq` pathway nvfp4 For any PR / issue related to NVFP4 support ready When a PR is ready for review
#2662 opened Apr 27, 2026 by brian-dellabetta Collaborator Loading…
1 task done
17
Add MixFP4A16 quantization recipe support enhancement New feature or request fp8 For any issue / PR related to FP8 support Refactor Code cleanup and/or improvements to existing features two-reviews When a PR requires two reviews
#2657 opened Apr 27, 2026 by revollllt Loading…
11
[Model] DeepSeekV4 quality-failed
#2655 opened Apr 26, 2026 by kylesayrs Collaborator • Draft
7
Transformers v5 needs-rebase
#2647 opened Apr 24, 2026 by kylesayrs Collaborator • Draft
14
[do not land] GPTQ actorder regression test suite awq For any issue / PR related to AWQ support fp8 For any issue / PR related to FP8 support gptq For any PR / issue related to GPTQ support llama For any PR / issue related to Llama herd support qwen For any PR / issue related to Qwen support w4a16
#2643 opened Apr 22, 2026 by HDCharles Collaborator • Draft
3 tasks
23
[AWQ] Seed grid search with identity baseline + fail fast on non-finite loss awq For any issue / PR related to AWQ support enhancement New feature or request ready When a PR is ready for review Refactor Code cleanup and/or improvements to existing features two-reviews When a PR requires two reviews
#2635 opened Apr 21, 2026 by juju812 Loading…
2 of 3 tasks
32
add example of w8a8fp8 for qwen3.5 documentation Improvements or additions to documentation enhancement New feature or request fp8 For any issue / PR related to FP8 support qwen For any PR / issue related to Qwen support two-reviews When a PR requires two reviews
#2631 opened Apr 20, 2026 by zhangxin81 Loading…
12
Adding test_group to lm-eval configs enhancement New feature or request fp8 For any issue / PR related to FP8 support nvfp4 For any PR / issue related to NVFP4 support two-reviews When a PR requires two reviews w4a16
#2623 opened Apr 16, 2026 by debroy-rh Loading…
15
test gptq issue [not for land] enhancement New feature or request gptq For any PR / issue related to GPTQ support nvfp4 For any PR / issue related to NVFP4 support quality-failed
#2617 opened Apr 14, 2026 by HDCharles Collaborator Loading…
10
[not for land] DDP regression tests awq For any issue / PR related to AWQ support documentation Improvements or additions to documentation enhancement New feature or request llama For any PR / issue related to Llama herd support quality-failed qwen For any PR / issue related to Qwen support
#2613 opened Apr 13, 2026 by HDCharles Collaborator Loading…
4 tasks done
17
fix: support transformers >= 5.0 (TORCH_INIT_FUNCTIONS fallback) bug Something isn't working qwen For any PR / issue related to Qwen support two-reviews When a PR requires two reviews w4a16
#2608 opened Apr 12, 2026 by quivent Loading…
8
[oneshot] clean offload_dir during post-processing
#2605 opened Apr 10, 2026 by brian-dellabetta Collaborator • Draft
3 tasks
4
[docs] deepseek v3.2 docs documentation Improvements or additions to documentation ready When a PR is ready for review
#2602 opened Apr 10, 2026 by brian-dellabetta Collaborator Loading…
18
Migrate to pyproject.toml and use dependabot for automatic handling quality-failed
#2600 opened Apr 10, 2026 by dsikka Collaborator • Draft
20
[Refactor] Refactor splits to only use the "calibration" split (#2551) ready When a PR is ready for review Refactor Code cleanup and/or improvements to existing features two-reviews When a PR requires two reviews
#2589 opened Apr 8, 2026 by arpitkh101 Loading…
1
28
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.