UPDATE JAN 13TH, 2026Quick update on my Capitan Conditioner Pack, original post <a target="_blank" rel="ugc" href="https://www.reddit.com/r/StableDiffusion/comments/1q9xdu7/conditioning_enhancer_qwenzimage_postencode_mlp/">here</a> if you missed it.The basic Conditioning Enhancer is unchanged (just added optional seed for reproducibility).New addition: Capitan Advanced Enhancer – experimental upgrade for pushing literal detail retention harder.It keeps the same core (norm → MLP → blend → optional attention) but adds:<ul><li>detail_boost (sharpens high-frequency details like textures/edges)</li><li>preserve_original (anchors to raw embeddings for stability at high mult)</li><li>attention_strength (tunable mixing – low/off for max crispness)</li><li>high_pass_filter (extra edge emphasis)</li></ul>Safety features like clamping + residual scaling let you crank mlp_hidden_mult to 50–100 without artifacts.Best use: Stack after basic, basic glues/stabilizes, advanced sharpens literally. Start super low strength (0.03–0.10) on advanced to avoid noise.Repo : <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer">https://github.com/capitan01R/Capitan-ConditioningEnhancer</a> Install via Comfyui Manager or git clone.Also qwen_2.5_vl_7b supported node is released. (usually used for Qwen-edit-2511), you can just extract to your custom nodes: <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer/releases/tag/vl7b_enhancer_2.0">latest release</a>Full detailed guide is available in the repo!!Full examples and Grid examples are available for both basic and advanced nodes in the repo files <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer/tree/main/capitan_enhancer_compare_examples">basic</a> &amp; <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer/tree/main/capitan_enhancer_compare_examples/advanced_basic">advanced</a>Let me know how it performs for you!Thanks for the feedback on the first version, appreciate it!!<hr />Just released Capitan Conditioning Enhancer – a lightweight ComfyUI custom node that improves the 2560-dim embeddings coming from Qwen3-4B (and now also Qwen 2.5 VL 7B).It sits between text encoder and KSampler and helps with: • better coherence &amp; scene unity • stronger prompt adherence • more consistent mood &amp; detail retentionWhat it does (in brief):<ul><li>Per-token normalization (usually keep on)</li><li>2-layer MLP refiner (starts very subtle)</li><li>Optional 8-head self-attention (helps distant prompt parts connect)</li></ul>Quick-start settings Everyday / stable look Strength: 0.00 – 0.10 Normalize: True Self-Attn: True MLP hidden mult: 2–4Detail / literal stack (my favorite)<ol><li>Glue node: Strength 0.05, Self-Attn True, Mult 2–4</li><li>Detail node: Strength -0.10, Self-Attn False, Mult 40–50</li></ol>Installation git clone <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer.git">https://github.com/capitan01R/Capitan-ConditioningEnhancer.git</a> into ComfyUI/custom_nodes/ → restart ComfyUIOR<ul><li>install it via Comfyui manager lookup: "Capitan-ConditioningEnhancer"</li></ul>Qwen 2.5 VL 7B support → grab the release here: <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer/releases/tag/qwen_2.5_vl_7b">here</a>Examples (grids + drag-and-drop workflows) added to repo: <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer">here</a>Reddit thread: <a target="_blank" rel="ugc" href="https://www.reddit.com/r/StableDiffusion/comments/1q9xdu7/conditioning_enhancer_qwenzimage_postencode_mlp/">here</a>Let me know what settings work best for you or if you run into any issues!Happy prompting!GitHub: <a target="_blank" rel="ugc" href="https://github.com/capitan01R/Capitan-ConditioningEnhancer">https://github.com/capitan01R/Capitan-ConditioningEnhancer</a>

Capitan Conditioning Enhancer – Better Prompt Adherence for Z-Image Workflows/Qwen-edit-2511

bug fixes added smarter memory management

Files with SHA256: 44c7763796d5d914eba162ddc558fea2ba8035fb1f2320675b583e05d64324b3