<p><strong>The plain Qwen Image 2.1 text-to-image graph: your prompt, your canvas, an optional LoRA stack, a PNG.</strong> Core Qwen nodes plus my free AusBoss nodes, nothing else.</p><p>New video: <a target="_blank" rel="ugc" href="https://youtu.be/GociGWQQDSs">I threw a ton of edits at Qwen Image 2.1, FLUX 3 Klein and FIBO 2</a></p><p>The first run renders the night bus-stop portrait from the gallery at 1536 × 1536 in 25 steps, in about 13 seconds on an RTX 5090.</p><h3 id="what's-inside">What's inside</h3><ul><li><p><strong>LoRA stack</strong>: starts empty. Add only adapters made for Qwen Image 2.1; trigger words of the rows you switch on join your text by themselves.</p></li><li><p><strong>Canvas</strong>: Latent Size picks the size (1536 × 1536 here) and drives the core empty latent. Keep its 32-pixel snap.</p></li><li><p><strong>Prompt</strong>: a Text node into Text Encode Qwen Image 2.1, which handles the positive and negative conditioning together.</p></li><li><p><strong>Seed</strong>: starts on Random, so every run after the first gives a new picture. Switch it to Fixed to keep one.</p></li><li><p><strong>Save</strong>: a PNG with the workflow embedded, plus a render timer and the Workflow Note card with download buttons.</p></li></ul><h3 id="quick-start">Quick start</h3><ol><li><p>Update ComfyUI. You need a build with <strong>Text Encode Qwen Image 2.1</strong> (mid-September 2026 or newer; I tested on 0.37.0).</p></li><li><p>Install <strong>AusBoss nodes</strong> from ComfyUI-Manager (search "AusBoss", registry id <code>ausboss-nodes</code>) and restart. You need 2.2.0 or newer. Manager installs the newest.</p></li><li><p>Drag the workflow in and download the three files from the card into the folders it names. No input image needed.</p></li><li><p>Write a prompt and queue.</p></li></ol><h3 id="models">Models</h3><ul><li><p><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/diffusion_models/qwen_image_2.1_int8_convrot.safetensors">qwen_image_2.1_int8_convrot.safetensors</a> → <code>models/diffusion_models</code> · 7.26 GB</p></li><li><p><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/text_encoders/qwen3vl_8b_int8_convrot.safetensors">qwen3vl_8b_int8_convrot.safetensors</a> → <code>models/text_encoders</code> · 9.35 GB</p></li><li><p><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/vae/qwen_image_2.1_vae_bf16.safetensors">qwen_image_2.1_vae_bf16.safetensors</a> → <code>models/vae</code> · 676 MB</p></li></ul><p>BF16 versions are on the same page: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Qwen-Image-2.1">Comfy-Org/Qwen-Image-2.1</a></p><h3 id="custom-nodes">Custom nodes</h3><p>Only <a target="_blank" rel="ugc" href="https://github.com/ausboss/ComfyUI-AusBoss"><strong>ComfyUI-AusBoss</strong></a>, 2.2.0 or newer (Manager installs it): free and open source. Text Encode Qwen Image 2.1 and the sampler are core ComfyUI. You don't need a Qwen wrapper pack.</p><h3 id="settings-that-worked-and-what-didn't">Settings that worked, and what didn't</h3><ul><li><p><strong>Sampling:</strong> 25 steps, CFG 1, Euler / simple, denoise 1: the starting values of the official template. 40 steps gives a more complete render. The negative prompt has no effect at CFG 1.</p></li><li><p><strong>Canvas:</strong> 1024 × 1024 is a fine start; try 2048 × 2048 for a larger render. The sample runs at 1536 × 1536; the rooftop shot in the gallery uses 1152 × 2048.</p></li><li><p><strong>Transparent output:</strong> ask explicitly for an RGBA image, a transparent background and an alpha channel, and keep PNG selected so the transparency survives saving.</p></li></ul><h3 id="speed">Speed</h3><p>Measured on an RTX 5090 (32 GB) with the INT8 files and the models loaded: about 13 s at 1536 × 1536 or 1152 × 2048, 25 steps.</p><p>I haven't tested smaller cards yet. ComfyUI offloads what doesn't fit, so expect it to run slower rather than fail.</p><h3 id="about-the-gallery">About the gallery</h3><p>All three portraits are straight outputs of this workflow; drag a PNG into ComfyUI to load its exact settings. The night bus stop is the default prompt (seed 7812); the rooftop at golden hour (seed 6311) and the photo booth (seed 6101) are other prompts. The cover is the bus-stop render plus a screenshot of the workflow's own Prompt card.</p><h3 id="license-and-credits">License &amp; credits</h3><p>Qwen Image 2.1 is © Alibaba Qwen, released under the Qwen Research License: <strong>non-commercial use only</strong>; commercial use needs a separate license from Qwen. The workflow and the AusBoss nodes are free.</p><h3 id="changelog">Changelog</h3><ul><li><p><strong>v1.0</strong> (2026-09-25): first release.</p></li></ul><p>Questions or bugs: <a target="_blank" rel="ugc" href="https://github.com/ausboss/ComfyUI-AusBoss/issues">GitHub issues</a> or the comments here. Post what you make with it; I read everything. Videos on <a target="_blank" rel="ugc" href="https://www.youtube.com/@ausbossdev">YouTube @ausbossdev</a>. I post new workflows on <a target="_blank" rel="ugc" href="https://x.com/Zanzibased">X @Zanzibased</a> and <a target="_blank" rel="ugc" href="https://github.com/ausboss">GitHub</a>.</p>

Files with SHA256: e961330c9059428bdc92b83fbc0957419854eebafaa0a292de3dc7fe3efac737