A Hugging Face community response advises ComfyUI users not to manually concatenate sharded .safetensors files such as part1of4.safetensors and part2of4.safetensors, and instead to start from the Comf…
A developer published a census of their homelab, documenting 41 Docker containers and 9 LXC containers spread across four machines, including a single 6 GB RTX 2060 that handles most AI routing decisi…
Japanese AI practitioners are increasingly embedding models inside existing tools rather than chatting with them, according to a weekly roundup of posts on X. Developers shinshin86 and rotejin used Cl…
Comfy API is now live for all users on a paid Comfy plan, letting teams deploy ComfyUI workflows as autoscaling API endpoints without rebuilding their environment elsewhere. The service, run from the …
A developer's guide compares leading AI image generators, concluding that Google's Nano Banana Pro (Gemini 3 Pro Image) is the best default for most users because it accepts conversational scene descr…
Dynaum released agdog, an open-source terminal monitor that fuses htop and nvitop by grouping CPU, RAM, GPU and VRAM usage by the AI agent that owns each process. agdog installs via Homebrew (brew ins…
Nvidia's RTX 5060 is usable for current PyTorch and image-generation stacks, but older packaged environments such as Fooocus's official portable still default to torch==2.1.0 and xformers==0.0.23, cau…
A developer has released EverSpark Forge, an open-source, self-hostable modular AI creation and orchestration system that unifies AI models, workflows, execution backends, runtime environments, and st…
ComfyUI launched Comfy Router, a unified API that lets its 3M+ users run generative media models such as Seedance 2.5, MiniMax H3, Nano Banana Pro, ChatGPT Images 2.5, KlingAI 3.0 and Flux 3 across pr…
Comfy launched Comfy Router on its Developer Platform, a single API that gives developers access to frontier image, video, 3D, and audio models from providers including fal, Runware, WaveSpeed, and Hi…
ComfyUI v0.36.0 ships optimizations that make the MiniMax H3 video VAE encode up to ~2.2x faster and decode ~1.4-2.7x faster, cutting a 1344x768, 129-frame encode-and-decode round trip from 24.3 to 12…
Alibaba's Qwen released Qwen-Image-2.1, a 7B-parameter image generation and editing model that combines a 32-layer single-stream DiT with a Qwen3-VL 8B text encoder and a 64-channel RGBA VAE, but ship…
Alibaba's Qwen team published Qwen-Image 2.1, a 7 billion parameter diffusion model that unifies text-to-image generation, instruction-based image editing, and native RGBA transparency in a single che…
A hands-on review of Qwen-Image 2.1 found the 7 billion parameter model produced a genuinely transparent RGBA phoenix image on the first attempt and arranged ten separately generated furniture pieces …
Alibaba's Qwen team released Qwen-Image-2.1, a 7B-parameter open-weight image generation and editing model that is now natively supported in ComfyUI, with weights published on Hugging Face. The model …
Livio_dev released Celestium Intel Cleaner V4 AUTO Background, a Windows optimization utility for Intel Arc, Iris Xe, and UHD GPUs that performs automatic background cleaning, shader cache reset, VRAM…
A developer testing NovelAI's V5 image generation API found that the params_version field — which third-party tools set inconsistently to 4, 3, 1, or 0 — produces pixel-for-pixel identical images when…
A developer verified that NovelAI's nai-diffusion-5-full and nai-diffusion-5-curated model IDs work through the ComfyUI_NAIDGenerator node, despite a README caveat that they had not been confirmed aga…
A developer measured NovelAI Diffusion V5's Anlas credit costs on an Opus subscription via the API, finding that generations up to Normal size (832×1216) at 28 steps cost 0 Anlas while Large (1024×153…
Tencent's Zhuque Lab released AI-Infra-Guard, an open-source security scanner for AI systems that fingerprints services such as Ollama, vLLM and ComfyUI, checks them against more than 1,600 known CVEs…