Search results

Int8 slower on GTX 1660 Super after 0.2.19

Status: Open.

#75 In Comfy-Org/comfy-kitchen;

· bakapotatolord opened on Jul 14, 2026

ComfyUI crashes using Flux Klein 9b with Triton backend - dequantize_nvfp4_kernel_tl

Status: Open.

#71 In Comfy-Org/comfy-kitchen;

· waltercool opened on Jul 10, 2026

INT8 matmul autotune uses NVIDIA-tuned configs on AMD RDNA

Status: Open.

#68 In Comfy-Org/comfy-kitchen;

· liminfei-amd opened on Jul 10, 2026

Memory access fault using Krea 2 INT8 CONVROT with Triton backend on AMD gfx1151

Status: Open.

#64 In Comfy-Org/comfy-kitchen;

· waltercool opened on Jul 7, 2026

Fatal access violation (0xC0000005) running int8_tensorwise model on ROCm Windows (gfx1101 / RDNA3)

Status: Open.

#61 In Comfy-Org/comfy-kitchen;

· kleintech opened on Jun 30, 2026

Python Python 3.13.11

Status: Open.

#45 In Comfy-Org/comfy-kitchen;

· saykorz opened on May 25, 2026

100 cpu when import

Status: Open.

#44 In Comfy-Org/comfy-kitchen;

· qwegjx opened on May 24, 2026

FP8 E4M3 quantized models inference is slower than same model quantized in MXFP8

Status: Open.

#38 In Comfy-Org/comfy-kitchen;

· matalama80td3l opened on May 18, 2026

The --no-cuda argument can't be passed through pyproject.toml

Status: Open.

#31 In Comfy-Org/comfy-kitchen;

· yurivict opened on Mar 19, 2026

Clarification regarding the readme example

Status: Open.

#27 In Comfy-Org/comfy-kitchen;

· sayakpaul opened on Jan 29, 2026

Hope to support older CUDA version backends

Status: Open.

#21 In Comfy-Org/comfy-kitchen;

· love530love opened on Jan 15, 2026

RTX30xx FP8 feather matmul triton kernel?

Status: Open.

#16 In Comfy-Org/comfy-kitchen;

· phazei opened on Jan 7, 2026