Skip to content

Releases: tchung1970/sd-cli-cuda

Release list

v1.0.2

v1.0.2 Latest
Latest

Choose a tag to compare

View all tags
@tchung1970 tchung1970 released this 09 May 21:02
v1.0.2
f251411

What's Fixed

  • CUDA memory check now actually runs — broken #ifdef SD_USE_CUDA guards removed (newer GGML uses GGML_BACKEND_DL, which prevented GGML_USE_CUDA from ever being defined)
  • Clean exit on OOM — changed exit(1) to _Exit(1) to skip C++ destructors and prevent SIGABRT crash

What's New

  • Detailed VRAM error popup — when generation is blocked by low VRAM, a centered zenity dialog shows:
    • Free/total VRAM in GB and the 8GB requirement
    • All GPU processes consuming VRAM (PID, name, usage MB)
    • sudo pkill -f main.py tip to free VRAM from ComfyUI
  • wavespeed-desktop error banner shows CUDA Out of Memory! immediately (popup is fully detached from sd-cli stderr)
  • xdotool auto-installed via postinst for popup centering
Assets 3
Loading

v1.0.1

v1.0.1

Choose a tag to compare

View all tags
@tchung1970 tchung1970 released this 03 Jan 20:18
v1.0.1
4edb812

sd-cli-cuda v1.0.1

CUDA-accelerated Stable Diffusion backend for wavespeed-desktop.

What's New

  • CUDA memory check - Requires 8GB free GPU memory before running
    • Exits immediately with clean "CUDA Out of Memory!" message when memory is low
    • Prevents silent failures that result in empty/blank images
    • Displays properly in wavespeed-desktop as "Generation failed: CUDA Out of Memory!"
  • CUDA runtime dependencies - Package declares proper dependencies
    • Auto-installs libcudart12, libcublas12, libcublaslt12 when using apt

Installation

# Recommended: auto-installs CUDA dependencies
sudo apt install ./sd-cli-cuda_1.0.1_amd64.deb

# Or manually (requires CUDA libs already installed)
sudo dpkg -i sd-cli-cuda_1.0.1_amd64.deb

Requirements

  • Debian-based distro (Ubuntu 24.04 LTS recommended)
  • NVIDIA GPU with CUDA support (8GB+ VRAM free required)
  • NVIDIA drivers installed (nvidia-driver-580 or newer for CUDA 12+)

Troubleshooting

If you see "CUDA Out of Memory!", free up GPU memory by closing GPU-intensive apps:

sudo pkill -f 'main.py'  # ComfyUI
Loading

v1.0.0

v1.0.0

Choose a tag to compare

View all tags
@tchung1970 tchung1970 released this 28 Dec 16:55
v1.0.0
0433e4c

sd-cli-cuda v1.0.0

CUDA-accelerated Stable Diffusion backend for wavespeed-desktop.

Features

  • CUDA GPU acceleration for NVIDIA GPUs (e.g., RTX 4000 Ada Generation)
  • Auto-integration with wavespeed-desktop
  • Z-Image support for local image generation

Installation

sudo dpkg -i sd-cli-cuda_1.0.0_amd64.deb

Requirements

  • Debian-based distro such as Ubuntu 24.04 LTS
  • NVIDIA GPU with CUDA support such as RTX 4000 Ada Generation
  • NVIDIA drivers installed such as nvidia-driver-580 (CUDA 12+)

What's Included

  • sd-cli - Command-line image generation tool
  • Symlink at ~/.config/wavespeed-desktop/sd-bin/sd for wavespeed-desktop integration
Loading