ComfyUI
ComfyUI, node-based local stable diffusion / image-and-video generation interface. Runs entirely on your GPU.
Release history
ComfyUI 0.42.1
Fixes crashes in the MOSS image template workflow; Apply if that workflow has been unstable
ComfyUI 0.42.0
ComfyUI 0.42.0 adds Krea 3 and MOSS-image nodes, improves Qwen 3.x performance, fixes webcam and text generation cleanup, and expands ARM Linux support
ComfyUI 0.41.0
ComfyUI 0.41.0 adds ImageSDRToHDR and ReferenceMultiLatent nodes, Flux3-image support, and latent geometry introspection for advanced image generation workflows
ComfyUI 0.40.0
ComfyUI 0.40.0 adds 3D asset export, camera info nodes, metadata attachment to prompts, partner model integrations (Vidu, Kling, Crystal), and VRAM optimizations for LLMs and attention mechanisms
ComfyUI 0.39.0
Adds Minimax-H3 VAE support and higher-quality defaults for video encoding
ComfyUI 0.38.0
Improves Qwen 2.1 KV cache location logic. Compiles Qwen Image 2.1 transformer blocks. Model files can now contain attention implementation hints
ComfyUI 0.37.0
Auto-detects and enables --fast-disk when disk is fast. Ships Aimdo 0.5.5
ComfyUI 0.36.0
New model blueprints; AMD Windows VA quota bumped to 4TB
ComfyUI 0.35.0
Drops retiring Veo 2 and 3.0 models; Adds Recraft V4 Styles
ComfyUI 0.33.1
Fixes KSamplerAdvanced nested latent handling, adds Minimax Music 3 and Bria image tools, improves VRAM management across platforms
ComfyUI 0.32.0
ComfyUI 0.32.0 adds LTX 2.5 support, optimizes memory handling for H3 models, and improves nested tensor debugging for complex workflows
ComfyUI 0.31.0
Reduces pinned memory on Linux systems without swap partitions
ComfyUI 0.29.2
Frontend improvements and new API/partner nodes for extending ComfyUI's extensibility and integration capabilities
ComfyUI 0.29.0
Streams video transcode instead of buffering all frames in memory
ComfyUI 0.28.0
Fixes Qwen3-VL tokenizer crash with custom embeddings; Adds AGENTS.md documentation
ComfyUI 0.27.0
Adds int8 convrot model support
ComfyUI 0.26.0
ComfyUI adds Qwen3-VL text generation, SCAIL-2 multireference, Boogu image editing, Kling V3-Turbo, Luma Rays 3.2, and job cancellation API endpoints
ComfyUI 0.25.1
ComfyUI added Kling V3-Turbo model support through partner nodes for faster video generation workflows
ComfyUI 0.25.0
ComfyUI adds audio consolidation, 3D preview nodes, partner integrations (Bria, Krea, Gemini), and performance optimizations for inference workflows
ComfyUI 0.24.0
Fixes dtype handling for birefnet and dino3 models, upgrade if using those
ComfyUI 0.23.0
MediaPipe face detection, multi-threaded model loading for faster startup, OAuth 2.1 support, and logic nodes shipped to improve performance and API standardization
ComfyUI 0.22.0
Audio processing nodes now handle None inputs gracefully
ComfyUI 0.21.1
Added Flux2, HiDream-O1, and Claude nodes, fixed fp8 safetensors saving and LTXV video alignment, improved memory usage