<p><strong>Blend multiple references into one sharp, seamless visual masterpiece.</strong></p><p><strong>Who it's for:</strong> creators who want this pipeline in ComfyUI without assembling nodes from scratch. <strong>Not for:</strong> one-click results with zero tuning - you still choose inputs, prompts, and settings.</p><h3 id="open-preloaded-workflow-on-runcomfy">Open preloaded workflow on RunComfy</h3><p><a target="_blank" rel="ugc" href="https://www.runcomfy.com/comfyui-workflows/krea-2-multi-image-reference-editing-comfyui-workflow-ai-visual-merge?utm_source=civitai&amp;utm_medium=referral">Open preloaded workflow on RunComfy (browser)</a></p><p><strong>Why RunComfy first</strong><br />- <strong>Fewer missing-node surprises</strong> - run the graph in a managed environment before you mirror it locally.<br />- <strong>Quick GPU tryout</strong> - useful if your local VRAM or install time is the bottleneck.<br />- <strong>Matches the published JSON</strong> - the zip follows the same runnable workflow you can open on RunComfy.</p><p><strong>When downloading for local ComfyUI makes sense</strong> - you want full control over models on disk, batch scripting, or offline runs.</p><p><strong>How to use (local ComfyUI)</strong><br />1. Load inputs (images/video/audio) in the marked loader nodes.<br />2. Set prompts, resolution, and seeds; start with a short test run.<br />3. Export from the Save / Write nodes shown in the graph.</p><p><strong>Expectations</strong> - First run may pull <strong>large weights</strong>; cloud runs may require a <strong>free RunComfy account</strong>.</p><hr /><h3 id="overview">Overview</h3><p>Using advanced Krea 2 engines, this workflow lets you blend multiple reference images into one refined visual output. You can mix identity, clothing, background, or product references for seamless, high-quality edits. The setup automates reference weighting, balancing styles without losing fidelity. Perfect for maintaining character consistency or testing creative scenarios. Enhance your projects with controlled image fusion for design, fashion, or marketing visuals. All elements work together smoothly to deliver crisp, realistic compositing results.</p><p><strong>Important nodes:</strong></p><h2 id="key-nodes-in-the-krea-2-multi-image-reference-editing-comfyui-workflow">Key nodes in the Krea 2 multi-image reference editing ComfyUI workflow</h2><h3 id="krea2editrebalance-(41)"><code>Krea2EditRebalance</code> (#41)</h3><p>Balances the influence of prompt and multiple references. Adjust <code>image1_tokens</code>, <code>image2_tokens</code>, <code>image3_tokens</code>, and optional <code>image4_tokens</code> to prioritize identity, wardrobe/product, or environment. If identity fidelity slips, increase the first image’s emphasis; if outfit or lighting underperforms, nudge their tokens up. This node is part of the Conditioning Rebalance custom package: <a target="_blank" rel="ugc" href="https://github.com/nova452/ComfyUI-Conditioning-Rebalance">nova452/ComfyUI-Conditioning-Rebalance</a>.</p><h3 id="unetloader-(37)"><code>UNETLoader</code> (#37)</h3><p>Loads the synthesis backbone, Krea-2 Turbo, which determines baseline style and capability. Keep this fixed for consistency across a campaign and change only when you intend a different rendering character. Official weights and notes: <a target="_blank" rel="ugc" href="https://huggingface.co/krea/Krea-2-Turbo">krea/Krea-2-Turbo</a> and <a target="_blank" rel="ugc" href="https://github.com/krea-ai/krea-2">krea-ai/krea-2</a>.</p><h3 id="ksamplerselect-(43)"><code>KSamplerSelect</code> (#43)</h3><p>Selects the sampling algorithm, which shapes texture and edge behavior. Use it to trade speed for crispness or softer gradients depending on your subject. Pair choices here with scheduler adjustments for predictable, repeatable improvements rather than wholesale resets.</p><h3 id="basicscheduler-(44)"><code>BasicScheduler</code> (#44)</h3><p>Controls step scheduling and overall denoise strength. Lower total refinement can keep more of the reference look; higher refinement pushes the model to reinterpret content more strongly. Tune this in tandem with sampler choice and conditioning emphasis to reach the desired balance of faithfulness and creativity. For general guidance on sigma scheduling concepts, see diffusion scheduling research like Karras et al. on arXiv.</p><h3 id="resolutionselector-(40)"><code>ResolutionSelector</code> (#40)</h3><p>Sets aspect ratio and size that feed the latent canvas. Match resolution to subject scale and output use (portrait, product hero, or scene). Larger sizes can resolve more texture but may require increased refinement to stabilize details.</p><p>…</p><h3 id="notes">Notes</h3><p><strong>Krea 2 Multi-Image Reference Editing ComfyUI Workflow | AI Visual Merge</strong> - see <a target="_blank" rel="ugc" href="https://www.runcomfy.com/comfyui-workflows/krea-2-multi-image-reference-editing-comfyui-workflow-ai-visual-merge?utm_source=civitai&amp;utm_medium=referral">RunComfy page</a> for the latest node requirements.</p>

Krea 2 Multi-Image Reference Editing ComfyUI Workflow | AI Visual Merge

<p>Initial release - Krea-2-Multi-Image.</p>

Files with SHA256: 44c7f89e5d65c4d1ebf794d13cb5b59fb7a48d2fbc6f9ded0f0d09082f2edbfc