ComfyUI Releases
111 releases of Comfy-Org/ComfyUI
- v0.3.62
This release focuses on workflow permissions, new portable installation links, and support for the Rodin3D Gen-2 model in API nodes.
Sep 30, 2025
- v0.3.61
This release introduces the Rodin3D Gen-2 API node, adds new audio nodes, and extends V3 schema support to many other nodes. It also adds support for Ascend NPU.
Sep 30, 2025
- v0.3.60
This release introduces extensive support for the Hunyuan image model, including distilled and refiner variants, along with V3 schema conversions for various API nodes like Kling, Wan, and Cosmos. It also adds fp8 ops support for specific hardware and updates the frontend.
Sep 23, 2025
- v0.3.59
A minor release focused on adding new API integrations, specifically introducing the ByteDanceSeedream (4.0) node, along with a template update for the wiki.
Sep 10, 2025
- v0.3.58
This release adds support for Hunyuan 3D 2.1 and addresses several AMD-specific optimizations, including bf16 VAE support. It also introduces StabilityAI Audio API nodes.
Sep 10, 2025
- v0.3.57
A major refactoring release focused on converting various node types to the V3 schema, including Stable Cascade, Video, and Primitive nodes. It also adds USO/UXO style reference support.
Sep 4, 2025
- v0.3.56
A performance-oriented release aimed at reducing RAM usage on Windows systems through internal optimizations.
Aug 30, 2025
- v0.3.55
This release expands support for Wan 2.2 models, adding specific support for the 5B fun control and inpaint models. It also introduces utility nodes for extending videos and cutting latents.
Aug 29, 2025
- v0.3.54
A maintenance release that updates the nodes_model_patch.py file, likely to ensure compatibility or fix issues with model handling.
Aug 28, 2025
- v0.3.53
This release introduces audio encoding capabilities with wav2vec2 and fixes critical crashes in EasyCache and LazyCache. It also adds preliminary support for Wan 2.2 S2V (Sound to Video) model.
Aug 28, 2025
- v0.3.52
This release focuses on V3 schema migration for String nodes and implements EasyCache and LazyCache for improved performance. It also adds support for Qwen Diffsynth controlnets.
Aug 23, 2025
- v0.3.51v0.3.51 Qwen Image Edit model
This release marks the implementation of the Qwen Image Edit model and adds support for various new API nodes including Minimax, GPT-5, and Vidu.
Aug 20, 2025
- v0.3.50
This release adds support for the Qwen Image model, introduces async API nodes, and fixes issues with multi-dimensional latents and AMD GPUs.
Aug 13, 2025
- v0.3.49
Focuses on device placement accuracy and introduces initial support for the Qwen Image model alongside a new Veo3 video generation node.
Aug 5, 2025
- v0.3.48
Updates ComfyAPI Core and introduces partial execution support alongside the V3 Node Schema Definition.
Aug 2, 2025
- v0.3.47
Optimizes memory usage for Wan2.2 models, specifically targeting the 5B i2v model and VAE operations.
Jul 30, 2025
- v0.3.46
Adds support for Wan2.2, XPU, and replaces torchaudio with pyav for audio handling.
Jul 28, 2025
- v0.3.45
Enhances training capabilities with async node functions and multi-dataset support, plus introduces the SA-Solver sampler.
Jul 21, 2025
- v0.3.44
Improves custom node management and adds new samplers, including Moonvalley nodes and image manipulation tools.
Jul 8, 2025
- v0.3.43
Adds Flux Kontext and Omnigen 2 models with fixes for memory estimation and controlnet compatibility.
Jun 27, 2025