最新资讯

官方博客 Doubling LLM token rates in ComfyUI ComfyUI runs a variety of large language models (LLMs) locally for enhancing prompts, writing stories, describing images, and even basic chatting. Recently, we rewrote the hot path of that text generation with a specific focus on the Qwen3.x family of LLMs. Th… rattus128 2026-11-12 官方博客 Seedream 5.5 is now available via Partner Nodes Seedream 5.5 is now available in ComfyUI via Partner Nodes. It is a multimodal model: it generates from text, and it edits from images.This release moves in five directions. Transparent RGBA generation. Precise localized editing that holds up across multiple r… Rob 2026-11-11 版本发布 ComfyUI v0.45.0 What's Changed Bump comfyui-frontend-package to 1.58.5 by @comfy-pr-bot in Add better error for when pixel_space VAE wrongly selected. by @comfyanonymous in Make SeedVR2 VAE slice length adaptive to free memory by @kijai in [Partner Nodes] feat(Pika): add Pika… ComfyUI 2026-11-11 官方博客 MiniMax H3 Image is now available in ComfyUI MiniMax H3 Image is now supported natively in ComfyUI. Open weights, six sampling steps, and output at 4096×4096. One model covers text-to-image and reference-image editing. It runs on your own hardware, and it is also available via Partner Nodes.Model Introdu… Rob 2026-11-11 官方博客 The little things that shipped in October October in ComfyUI brought a collection of new model families and Partner Nodes, but those weren’t the only things new to ComfyUI. Alongside these model drops, we shipped a long list of smaller quality-of-life changes to the graph, the canvas and the tools aro… Joshua Jones 2026-11-10 版本发布 ComfyUI v0.44.0 What's Changed [Partner Nodes] feat(Patina): add Patina texture nodes by @bigcat88 in [Partner Nodes] feat(VEED): add VEED Dubbing node by @bigcat88 in Update workflow templates to v0.11.98 by @comfyui-wiki in Switch pylint py version to 3.12 by @comfyanonymou… ComfyUI 2026-11-04 版本发布 ComfyUI v0.43.0 What's Changed model_prefetch: syncrhonize all graph cleanups by @rattus128 in Add rope scaling support to MOSS-image by @kijai in [Partner Nodes] Renew partner-node sign-in tokens in API nodes by @dante01yoon in [Partner Nodes] feat(HeyGen): add HeyGen Voice … ComfyUI 2026-10-27 版本发布 ComfyUI v0.42.1 Fix some crashing issues a few users experienced when running the MOSS image template workflow. ComfyUI 2026-10-19 版本发布 ComfyUI v0.42.0 What's Changed [Partner Nodes] feat(Krea): add Krea 3 Image node by @bigcat88 in Update workflow templates to v0.11.85 by @comfyui-wiki in Use FLUX 3 Image in node display labels by @MaanavD in Better documentation about the comfy quant format. by @comfyanonym… ComfyUI 2026-10-19 官方博客 Higgsfield promised to open source their entire platform. They didn’t, so let’s help them. Allyson Toy 2026-10-16 版本发布 ComfyUI v0.41.0 What's Changed Add ImageSDRToHDR node. by @comfyanonymous in feat: Add a core route for the default workflow (CORE-340) by @alexisrolland in Add ReferenceMultiLatent node. by @comfyanonymous in Disable the LoRA stack nodes until the frontend supports DynamicGr… ComfyUI 2026-10-16 官方博客 Krea 3 is now available via Partner Nodes Today, Krea 3 is available in ComfyUI through Partner Nodes.Krea 3 is the new image model from Krea, built for generation and editing driven by references. You start from something real: a product shot, a person, a room, or a stack of visual references. You de… Rob 2026-10-16 官方博客 FLUX 3 Image Klein in ComfyUI: 5.6B Open Weights FLUX 3 Image Klein is the compact, open-weights model in Black Forest Labs’ FLUX 3 family. It draws on learning shared across the family’s image, video, and audio models, and it keeps the control features of the larger FLUX 3 Image model, bounding boxes includ… Rob 2026-10-16 版本发布 ComfyUI v0.40.0 What's Changed [Server.py] Update CSP to allow `blob:` and `data:` by @S0yKaf in Add comfy_attention and AttentionTensorContainer support to rest of models by @comfyanonymous in Implement the asset export API locally (CORE-454) by @jtydhr88 in Report 3D models… ComfyUI 2026-10-14 官方博客 Nvidia VFX Nodes for Video Upscaling, Frame Generation and SDR-to-HDR, for VFX Professionals Nvidia’s RTX video effects now run inside ComfyUI as Convert Image Color Space node, EXR preview in Load Image and Save Image (Advanced) nodes.Preview & Download EXRsEXRs also get a major quality-of-life improvement: they are now easier to inspect and to downl… Rob 2026-10-14 官方博客 How I Generated Live Video with MiniMax H3 on a Single GPU Bryan Wade 2026-10-09 官方博客 Gemini Nano Banana 2.1 is now available via Partner Nodes Thanks for reading ComfyUI Newsletter! Subscribe for free to receive new posts and support my work.Gemini Nano Banana 2.1 is Google’s image generation and editing model, and the successor to Nano Banana 2. It is built to balance price and performance. Give it … Rob 2026-10-07 版本发布 ComfyUI v0.39.0 What's Changed Support LynnReal light Minimax-H3 vae by @kijai in Update embedded docs to v0.5.13 by @comfyui-wiki in Use higher quality defaults for Save Video encoding. by @comfyanonymous in feat: add DynamicGroup widget input by @jaeone94 in fix(assets): do… ComfyUI 2026-10-06 官方博客 Creating the story of YUI with 852話 Hakoniwa and Comfy Agent A girl, a timer, and a wall.YUI is the first animated short film created with Jo Zhang 2026-10-02 官方博客 Comfy Agent: The First Agent for Craft Today we are excited to introduce Jo Zhang 2026-10-02 官方博客 Comfy API Is Live: Deploy ComfyUI Workflows as Production APIs Today Comfy API is live for everyone on a paid Comfy plan.Turn a workflow you built in ComfyUI into the engine behind an app, tool, or service. Package the workflow with its custom nodes, LoRAs, models, and Python dependencies, then deploy it as an autoscaling… Jacob Segal 2026-10-01 版本发布 ComfyUI v0.38.0 What's Changed Improve qwen 2.1 KV cache location logic by @rattus128 in Compile Qwen Image 2.1 transformer blocks by @rattus128 in Allow model files to contain which attention should be used for each block. by @comfyanonymous in JsonExtractString can now hand… ComfyUI 2026-09-30 官方博客 Introducing Comfy Router: One API for Frontier Media Models Today Day one access includes Seedance 2.5, MiniMax H3, Nano Banana Pro, GPT Image 2, Kling, and Black Forest Labs, with more coming. These are the same models 3M+ ComfyUI users already use, now behind one API. Router runs on the Developer Platform.Router runs… Matt Miller 2026-09-23 官方博客 Making the MiniMax H3 Video VAE 2x Faster The MiniMax H3 video VAE now encodes up to ~2.2x faster and decodes ~1.4-2.7x faster. Together that can roughly halve the time a video workflow spends in the VAE: a 1344x768, 129-frame encode and decode round trip drops from 24.3 to 12.7 seconds.What changed, … Jukka Seppänen 2026-09-23 版本发布 ComfyUI v0.37.0 What's Changed Aimdo 0.5.5 + Auto-detect and enable --fast-disk when the disk is fast (CORE-440) by @rattus128 in [Partner Nodes] feat(OpenAI): add transparent background support for GPT Image 2 by @bigcat88 in Add CFG control to YuE2 Generate Music node. by @… ComfyUI 2026-09-21 官方博客 Qwen-Image-2.1 in ComfyUI: Open-Weight Image Generation and Editing, Now with Transparency A new open-source image editing model has finally arrived. Qwen-Image-2.1 is now supported natively in ComfyUI. Open weights, 7B, and it generates a real alpha channel. Rob 2026-09-20 版本发布 ComfyUI v0.36.0 What's Changed Add new model blueprints and reorganize subgraph categories by @comfyui-wiki in main: bump the AMD Windows VA quota to 4TB (CORE-409) by @rattus128 in [Partner Noes] feat(OpenRouter): add Microsoft `mai-image-2.6` models by @bigcat88 in quant_op… ComfyUI 2026-09-16 版本发布 ComfyUI v0.35.0 What's Changed [Partner Nodes] chore(Google): drop retiring Veo 2 and Veo 3.0 models by @bigcat88 in [Partner Nodes] feat(Recraft): add V4 Styles by @bigcat88 in [Partner Nodes] feat(WAN): add WAN3-Prime model support by @bigcat88 in Bump comfyui-frontend-pack… ComfyUI 2026-09-10 官方博客 Forward Deployed Creatives Most Enterprise teams we talk to are not actually stuck on whether “genAI” can do the task. They’ve already proven that it can. Somebody on the team made something genuinely good, showed it around the studio, and everybody in the room got really excited about … Doug Hogan 2026-09-05 官方博客 Comfy H3 Sync Sound Challenge: The Winners The POM (Banodoco founder), Yachimat (animation and manga artist). Their scores were averaged and added to the Comfy team’s, and a fourth title, Built with MCP, was judged on its own track.Watch the Visual Frisson generated a large volume of H3 clips using the… Allyson Toy 2026-09-04