Search results

Add weight activation ordering for fp8 block

enhancementNew feature or requestNew feature or request

fp8For any issue / PR related to FP8 supportFor any issue / PR related to FP8 support

good first issueA good first issue for users wanting to contributeA good first issue for users wanting to contribute

Status: Open.

#2587 In vllm-project/llm-compressor;

· HDCharles opened on Apr 8, 2026

[Bug]: Using AWQModifier with kv_cache_scheme fails.

bugSomething isn't workingSomething isn't working

Status: Open.

#2584 In vllm-project/llm-compressor;

· siahuat0727 opened on Apr 8, 2026

[Bug]: GPTQ INT4 quantized Qwen3-vl-30b-a3b can not deploy by vllm with tensor-parallel-size=2

bugSomething isn't workingSomething isn't working

Status: Open.

#2582 In vllm-project/llm-compressor;

· freeneuro opened on Apr 8, 2026

Getting "KeyError: language_model.language_model.layers.0.mlp.experts.w2_weight_packed" while trying to infer AWQ quantized Qwen3.5-122B model using vllm

Status: Open.

#2568 In vllm-project/llm-compressor;

· Priya95715 opened on Apr 6, 2026

[Bug]: Encountered a type error related to offload.device.type when running kimi_k2_example.py

bugSomething isn't workingSomething isn't working

Status: Open.

#2563 In vllm-project/llm-compressor;

· NanPaiUncle3 opened on Apr 3, 2026

[Bug]: Cannot properly compress remaining layers of already compressed model (gpt-oss-20b)

bugSomething isn't workingSomething isn't working

Status: Open.

#2560 In vllm-project/llm-compressor;

· jlmckins opened on Apr 2, 2026

[Refactor] Remove PreprocessingFunctionRegistry

enhancementNew feature or requestNew feature or request

good first issueA good first issue for users wanting to contributeA good first issue for users wanting to contribute

RefactorCode cleanup and/or improvements to existing featuresCode cleanup and/or improvements to existing features

Status: Open.

#2552 In vllm-project/llm-compressor;

· kylesayrs opened on Apr 1, 2026

[Refactor] Refactor splits to only use the "calibration" split

enhancementNew feature or requestNew feature or request

good first issueA good first issue for users wanting to contributeA good first issue for users wanting to contribute

Status: Open.

#2551 In vllm-project/llm-compressor;

· kylesayrs opened on Apr 1, 2026

llmcompressor W8A8 is slower than FP16

Status: Open.

#2549 In vllm-project/llm-compressor;

· DarkenStar opened on Mar 31, 2026

GPTQ ActivationOrdering support for per-channel quantization

enhancementNew feature or requestNew feature or request

good first issueA good first issue for users wanting to contributeA good first issue for users wanting to contribute

gptqFor any PR / issue related to GPTQ supportFor any PR / issue related to GPTQ support

Status: Open.

#2524 In vllm-project/llm-compressor;

· HDCharles opened on Mar 26, 2026

[Bug]: Evaluate AWQ and GPTQ for Gemma

bugSomething isn't workingSomething isn't working

keep-open

Status: Open.

#2522 In vllm-project/llm-compressor;

· Jeevi10 opened on Mar 26, 2026

[Bug]: Fix ML3 FP8 documentation example

bugSomething isn't workingSomething isn't working

documentationImprovements or additions to documentationImprovements or additions to documentation

Status: Open.

#2515 In vllm-project/llm-compressor;

· kylesayrs opened on Mar 25, 2026