llm-compressor
Python Mediumvllm-project/llm-compressor
2,989 stars
473 forks
15 open issues
Active Apr 2026
Beginner-Friendly Issues 15
Issues tagged for new contributors
Add weight activation ordering for fp8 block
#2587 · Apr 8, 2026
enhancement good first issue fp8
[Bug]: Using `AWQModifier` with `kv_cache_scheme` fails.
#2584 · Apr 8, 2026
bug
bug
[Refactor] Remove `LogarithmicEqualizationModifier`
#2554 · Apr 1, 2026
enhancement good first issue Refactor
[Refactor] Remove `PreprocessingFunctionRegistry`
#2552 · Apr 1, 2026
enhancement good first issue Refactor
llmcompressor W8A8 is slower than FP16
#2549 · Mar 31, 2026
GPTQ ActivationOrdering support for per-channel quantization
#2524 · Mar 26, 2026
enhancement good first issue gptq
Refactor Activation Ordering across LLMC and CT
#2523 · Mar 26, 2026
enhancement good first issue gptq
[Bug]: Evaluate AWQ and GPTQ for Gemma
#2522 · Mar 26, 2026
bug
[GPTQ] Add support for block-wise quantization
#2520 · Mar 25, 2026
enhancement good first issue
[Bug]: Fix ML3 FP8 documentation example
#2515 · Mar 25, 2026
bug documentation
[Feature Request/Question]fused QKV linear layers
#2494 · Mar 20, 2026
enhancement
Consolidate Intermediate Offloading
#2490 · Mar 19, 2026
enhancement good follow-up issue Refactor
[Feature] Enable GPU offloading for reduced weight movement
#2480 · Mar 17, 2026
enhancement good first issue
[Bug]: when ignore specific module, bug happens
#2454 · Mar 9, 2026
bug