Skip to content
View 0xDELUXA's full-sized avatar
🎯
Focusing
🎯
Focusing
20 followers · 17 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: Quickdraw

Block or report 0xDELUXA

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
0xDELUXA/README.md

DELUXA

Systems-level contributor focused on PyTorch, AMD ROCm on Windows, and ML infrastructure.

Notable Contributions

RepositoryContribution
huggingface/transformersFix: Conditionally import `torch.distributed.fsdp` in `trainer_seq2seq.py`
Comfy-Org/ComfyUIEnable PyTorch Attention by default on gfx1200
Enable fp8 ops by default on gfx1200
pytorch/pytorch[elastic] Add Windows support for stdout/stderr redirects
Dao-AILab/flash-attention[ROCM] Fix windows issues (#2385)
triton-lang/triton[AMD] Stop lowering bf16 multiply to v_dot2_bf16_bf16
vosen/ZLUDAdocs: clarify unofficial HIP SDK column refers to AMD nightlies
huggingface/accelerateFix: Conditionally import `torch.distributed.algorithms.join` in `accelerator.py`
deepbeepmeep/Wan2GPUpdate `AMD-INSTALLATION.md`
Revise AMD installation guide
LykosAI/StabilityMatrixFix: Pin OneTrainer ROCm bitsandbytes wheel to 0.49.1
bitsandbytes-foundation/bitsandbytes[ROCm] Restore Wave64 warp size for all gfx9 targets

Auto-updated via GitHub Actions

Pinned Loading

  1. bitsandbytes_win_rocm bitsandbytes_win_rocm Public

    Forked from guinmoon/bitsandbytes_win_rocm

    Accessible large language models via k-bit quantization for PyTorch - a fork mainly for hosting prebuilt wheels for AMD users

    Python 12 1

  2. comfy-kitchen_win-rocm comfy-kitchen_win-rocm Public

    Forked from Comfy-Org/comfy-kitchen

    Fast kernel library for Diffusion inference with multiple compute backends. - HIP Backend Fork

    Python 7 1

  3. ComfyUI-DN_PatchFlashAttention ComfyUI-DN_PatchFlashAttention Public

    ComfyUI custom node to patch default attention with Flash Attention 2

    Python 8 2

  4. ComfyUI-DN_PatchVAEAttention ComfyUI-DN_PatchVAEAttention Public

    ComfyUI custom node to patch the default attention in VAE to a specific implementation

    Python 1 2

  5. comfy-int8-convrot-to-w4a8 comfy-int8-convrot-to-w4a8 Public

    Convert a ComfyUI int8_tensorwise ConvRot checkpoint to asym_w4a8_int8 without the original fp16 weights.

    Python 1

  6. ComfyUI-MemoryVisualization ComfyUI-MemoryVisualization Public

    Forked from kijai/ComfyUI-MemoryVisualization

    JavaScript