Search results

nvcc 13.1 ptxas miscompile on sm_90 drops createpolicy for cp.async cache_hint

Status: Open.

#44 In IST-DASLab/marlin;

· huangzhilin-hzl opened on May 1, 2026

Any plan on AMD ROCm platform support?

Status: Open.

#43 In IST-DASLab/marlin;

· charyang-ai opened on Oct 29, 2025

Potential Async Data Race in cp_async4_stream Usage - Request for Validation

Status: Open.

#42 In IST-DASLab/marlin;

· Zephyreeze opened on Jul 24, 2025

Question about the meaning of slice_count and slice_idx in global reduction condition

Status: Open.

#41 In IST-DASLab/marlin;

· ZhengChen6 opened on Jul 24, 2025

Issue with running kernel on blackwell GPUs (RTX 5090)

Status: Open.

#40 In IST-DASLab/marlin;

· huoshenlaile opened on Mar 10, 2025

Different batch size results to different result for gptq_marlin_gemm kernel.

Status: Open.

#38 In IST-DASLab/marlin;

· rooooc opened on Dec 30, 2024

Question about matrix A

Status: Open.

#37 In IST-DASLab/marlin;

· Zhao-Dongyu opened on Oct 23, 2024

Marlin vs gguf

Status: Open.

#36 In IST-DASLab/marlin;

· blap opened on Sep 13, 2024

comparison with FastTransformer for BatchSize=1

Status: Open.

#35 In IST-DASLab/marlin;

· matrix97317 opened on Sep 11, 2024

How to move the pointer of B and s in cuda kernel

Status: Open.

#34 In IST-DASLab/marlin;

· Eutenacity opened on Aug 22, 2024

comparison with SmoothQuant

Status: Open.

#32 In IST-DASLab/marlin;

· lxww302 opened on Jul 29, 2024

RuntimeError: CUDA error: an illegal instruction was encountered when runing test.py

Status: Open.

#30 In IST-DASLab/marlin;

· MekkCyber opened on Jul 2, 2024