═══════════════════════════════════════════════════════════════════════════════ QWEN EDIT FIGURE MAKER BY ALDNIKI ═══════════════════════════════════════════════════════════════════════════════ ─────────────────────────────────────────────────────────────────────────────── IDENTITY ─────────────────────────────────────────────────────────────────────────────── Base model Qwen-Image Generation mode Image Editing (instruction editing) Checkpoint Qwen-Image (original T2I, 2025-08-04) Category Visual style Author aldniki217 Source https://civitai.com/models/1908503 Provenance CivitAI · model 1908503 · version 2160163 ─────────────────────────────────────────────────────────────────────────────── WEIGHTS ─────────────────────────────────────────────────────────────────────────────── File aldniki_qwen_figure_maker_v01.safetensors Version v1.0 Size 562.7 MB Format SafeTensor / unknown precision SHA256 17fd4fb848e51fb6ffdd60fc5b093cbe83521ab6b5bc58c9b90ce6b37a324376 Published 2025-08-29 ─────────────────────────────────────────────────────────────────────────────── TRIGGER WORDS ─────────────────────────────────────────────────────────────────────────────── › "补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办" ─────────────────────────────────────────────────────────────────────────────── RECOMMENDED SETTINGS ─────────────────────────────────────────────────────────────────────────────── Taken verbatim from the author's description: LoRA strength 1.0 Sampler steps 8 CFG / guidance 1.0 Sampler Euler ─────────────────────────────────────────────────────────────────────────────── ADOPTION & POPULARITY ─────────────────────────────────────────────────────────────────────────────── Downloads 9,494 Rating 100% positive (951/952) Comments 0 Published 372 days ago Download rate 25.5/day since release Adoption rank #36 of 150 in this collection (High) Momentum rank #10 of 150 by download rate (Hot) Adoption tier is the percentile of total downloads within this collection; momentum is the percentile of downloads-per-day since release. Momentum is ranked rather than measured against a fixed rate, because this base model is itself new and every LoRA looks fast on an absolute scale. Both numbers are CivitAI's, read on the scrape date at the foot of this file. ─────────────────────────────────────────────────────────────────────────────── CHAINING / STACKING ─────────────────────────────────────────────────────────────────────────────── Stacks with others not stated The author gave no chaining guidance. The rules below are the general Qwen-Image ones, not anything specific to this LoRA. General Qwen-Image rules that apply here (see REFERENCE.txt): • Stack by wiring LoraLoaderModelOnly nodes in series on the MODEL line. This is the officially shipped pattern: template_qwen_Image_2512_360_lora chains Qwen- Image-2512-Lightning-4steps-V1.0-fp32 at 1.0 into qwen-360-diffusion-2512-int8-bf16-v2 at 1.0. • One accelerator LoRA maximum. Every official template loads exactly one Lightning/Turbo LoRA, and ComfyUI's Qwen-Image docs record that the fully-distilled checkpoint and the lightx2v 8-step LoRA are not usable together. • Match the accelerator to the checkpoint family: Qwen- Image-2512-Lightning on 2512, Qwen-Image-Lightning on the August base, the Edit-2509 subfolder LoRAs on Edit-2509, Qwen-Image- Edit-2511-Lightning on Edit-2511. lightx2v distilled and shipped these separately per family. • When any accelerator LoRA is in the chain, move the sampler to that LoRA's step count and cfg 1.0. Leaving cfg at 4.0 with a distilled LoRA is the most common bad-output cause. • Order used by the official templates: accelerator LoRA adjacent to the base loader, concept/style LoRA after it, then ModelSamplingAuraFlow, then the sampler (the 360-panorama template's exact order). • Keep the summed strength of concept/style LoRAs roughly in the 0.9-1.2 band (CivitAI article 18798). The accelerator LoRA sits outside that budget at its own 1.0. • Control LoRAs (qwen_image_union_diffsynth_lora.safetensors) chain like ordinary LoRAs and combine with an accelerator: the official 'Qwen- Image Union Control' template stacks the control LoRA with Qwen-Image- Lightning-4steps-V1.0, both at 1.0, sampling at 4 steps / cfg 1.0 / shift 3.1. • Be conservative stacking older Edit LoRAs on Edit-2511: the Qwen README says community LoRA effects were merged into the 2511 base, so those LoRAs can double-apply. • 20B MMDiT. Comfy-Org/Qwen-Image_ComfyUI split_files/diffusion_models (verified sizes): qwen_image_bf16.safetensors 40.86 GB, qwen_image_fp8_e4m3fn.safetensors 20.43 GB, qwen_image_fp8_hq 22.74 GB, qwen_image_fp8mixed 20.53 GB, qwen_image_nvfp4 19.77 GB. Same repo also carries non_official/ full-distill checkpoints qwen_image_distill_full_bf16 (40.86 GB) and qwen_image_distill_full_fp8_e4m3fn (20.4 ─────────────────────────────────────────────────────────────────────────────── PROS ─────────────────────────────────────────────────────────────────────────────── + Widely adopted — 9,494 downloads. [source: CivitAI stats] + Positively received — 951 thumbs-up, 1 down. [source: CivitAI stats] + Explicit trigger word(s) — 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办 — so the effect can be turned on and off from the prompt. [source: model metadata] + Author published concrete recommended settings. [source: description] ─────────────────────────────────────────────────────────────────────────────── CONS & CAVEATS ─────────────────────────────────────────────────────────────────────────────── (none recorded) ─────────────────────────────────────────────────────────────────────────────── COMPATIBILITY WARNING ─────────────────────────────────────────────────────────────────────────────── Mismatch is silent degradation, essentially never a hard error. Every Qwen- Image weight family resolves to the same ComfyUI model class (comfy.model_base.QwenImage, unet_config image_model "qwen_image", latent_format Wan21, built-in sampling shift 1.15 - comfy/supported_models.py), and the bf16/fp8 file sizes are identical across Qwen-Image, Qwen-Image-2512, Qwen-Image-Edit, Edit-2509 and Edit-2511. So a LoRA trained on any of them will key-match and load against any other. Where keys genuinely do not match, ComfyUI only logs `lora key not loaded: ` (comfy/lora.py:93) and continues; a shape mismatch logs `WARNING SHAPE MISMATCH WEIGHT NOT MERGED` (comfy/lora.py:498) and skips that tensor. Both are console warnings; the render still completes. Verified consequence: lightx2v re-distilled and re-released Lightning weights separately for Qwen-Image, Qwen-Image-Edit, Edit-2509, Edit-2511 and Qwen-Image-2512, so accelerators are family- specific in practice. The one documented case worse than mild degradation is precision-level, not family-level: bf16-trained LoRAs on the naive qwen_image_fp8_e4m3fn.safetensors base produce grid artifacts, stated outright in the ModelTC/Qwen-Image-Lightning README. Whether an Edit LoRA on the T2I base merely no-ops (no ref_latents conditioning is supplied) or a T2I LoRA on an Edit checkpoint transfers as a style push is inference from the architecture, not a sourced statement - see unverified. ─────────────────────────────────────────────────────────────────────────────── AUTHOR'S DESCRIPTION ─────────────────────────────────────────────────────────────────────────────── You can use the online workflow below. If the results are good, you can then try setting it up and running it locally. 👉 Click Here For Online WorkFlow As a fan benefit, you’ll receive 1,000 points upon registration, plus 100 points for daily logins. Enjoy a smooth experience with an RTX 4090 and 48GB of VRAM. Model Description This is an upgraded version of the previously released Kontext Figure Maker , trained on the Qwen Image Edit model, with results far, far better than the Kontext version ! This version can reliably produce realistic figures in a style similar to Nano Banana (now Gemini) . It also allows you to change the original image’s scene by switching prompts—see the prompt recommendations for details. And don’t forget the excellent Chinese support in the Qwen Image series. Recommended Settings - Sampler : Euler - LoRA Weight : 1 - CFG : 1.0 (Using the Qwen-Image-Lightning-8steps-V1.1 , acceleration LoRA weight 1.0) - Steps : 8 - Recommended to use high-resolution upscale: 1.5× scale. Denoising set to 0.5. Beta scheduler is recommended, with 8 steps, CFG 1.0, using the Qwen- Image-Lightning-8steps-V1.1 , acceleration LoRA weight 0.5. Recommended Prompt (In Chinese) You can also write your prompts in English. Most of the sentences and words in the prompt can be replaced, so feel free to explore! Trigger Word :补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办 Other Prompts : - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,在手办的后面,有一个比手办大的设计精美的二次元手办包装盒,包装盒上写 着精美设计的“千问QWEN”艺术字,包装右下角有正方形的红色标签,红色标签上写着“ALDNIKI”,“ALD”和“NIKI”各占标签的一行,包装盒上印 着图片形象,在包装盒后面还有一个AOC牌子的电脑屏幕,屏幕上显示着3d建模软件中手办的灰色3d模型,手办被摆放在一个半透明的展示盘上,强调手办的pvc 塑料质感,场景在一个手办生产工作室内,后面有3d打印机 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办被拿在一只真实的手里把玩,强调手办的pvc塑料质感,场景在一个手办生产工作室内 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办被放置在漫展摊位上进行售卖,后面有正在售卖的工作人员的身体 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办置于会发光的魔法法阵中央,法阵浮现半透明光线和神秘符文,手办表面 带有逼真PVC光泽,法阵散发蓝色和金色光芒在手办上投射动态光影,背景是古老手办工作室与奇幻元素结合的场景,桌面上散落着模型工具和魔法书籍,整体画面呈现 幻想实验室的神秘氛围。 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办被置于未来感浮空展示架上,周围漂浮着微型全息屏幕显示手办设计信息 ,光线从架顶射下形成逼真阴影,PVC表面细节清晰,背景是高科技手办展览厅,氛围冷色调科技感十足,同时有微型机械臂在旁辅助整理手办道具。 - (Use your imagination) 模型介绍 这是此前发布的 Kontext Figure Maker 的升级版本,基于 Qwen Image Edit 模型训练, 效果比Kontext版好的多得多得多 ! 此版本可以稳定地产出类似 Nano Banana (now Gemini) 风格的真实手办,同时此版本可以尝试切换提示词来更改原图的场景,详见提示词推荐。同时不要忘了Qwen Image系列的中文支持。 推荐设置 - 采样器 : Euler - LoRA 权重 : 1 - CFG : 1.0 (使用 Qwen-Image-Lightning-8steps-V1.1 , 加速LoRA权重1.0) - 步数 : 8 - 推荐使用高清放大 : 1.5倍. 降噪 0.5. Beta 调度器, 8步, CFG 1.0, 使用 Qwen-Image- Lightning-8steps-V1.1 , 加速LoRA权重0.5 推荐提示词 你同样可以使用英文来编写提示词, 提示词中的大部分句子和词语都可以被替换,请尽情探索! 触发词:补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办 其他提示词: - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,在手办的后面,有一个比手办大的设计精美的二次元手办包装盒,包装盒上写 着精美设计的“千问QWEN”艺术字,包装右下角有正方形的红色标签,红色标签上写着“ALDNIKI”,“ALD”和“NIKI”各占标签的一行,包装盒上印 着图片形象,在包装盒后面还有一个AOC牌子的电脑屏幕,屏幕上显示着3d建模软件中手办的灰色3d模型,手办被摆放在一个半透明的展示盘上,强调手办的pvc 塑料质感,场景在一个手办生产工作室内,后面有3d打印机 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办被拿在一只真实的手里把玩,强调手办的pvc塑料质感,场景在一个手办生产工作室内 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办被放置在漫展摊位上进行售卖,后面有正在售卖的工作人员的身体 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办置于会发光的魔法法阵中央,法阵浮现半透明光线和神秘符文,手办表面 带有逼真PVC光泽,法阵散发蓝色和金色光芒在手办上投射动态光影,背景是古老手办工作室与奇幻元素结合的场景,桌面上散落着模型工具和魔法书籍,整体画面呈现 幻想实验室的神秘氛围。 - 补全主体的躯体,保持姿势和面部特征不变,将主体变成一个写实的3d pvc手办,手办被置于未来感浮空展示架上,周围漂浮着微型全息屏幕显示手办设计信息 ,光线从架顶射下形成逼真阴影,PVC表面细节清晰,背景是高科技手办展览厅,氛围冷色调科技感十足,同时有微型机械臂在旁辅助整理手办道具。 - (发挥你的想象) ─────────────────────────────────────────────────────────────────────────────── EXAMPLE MEDIA ─────────────────────────────────────────────────────────────────────────────── 6 preview(s) in ./example_images/ — community results for this LoRA. Videos keep a matching _poster.jpg still frame. No prompts are recorded below: CivitAI's public API flags these posts as having generation metadata but does not return it, so the prompts cannot be scraped. Open the model page to read them. 01_image.jpeg (image, 1008x1008) 02_video.mp4 (video, 1228x1536) still: 02_poster.jpg 03_video.mp4 (video, 1024x1536) still: 03_poster.jpg 04_image.jpeg (image, 1248x1872) 05_video.mp4 (video, 1024x1536) still: 05_poster.jpg 06_image.jpeg (image, 1248x1872) ─────────────────────────────────────────────────────────────────────────────── HOW TO USE (COMFYUI) ─────────────────────────────────────────────────────────────────────────────── 1. Update ComfyUI first - Qwen-Image support is native, but each new family member (Edit-2509, Edit-2511, 2512) needed a fresh release, so an old build will not detect the checkpoint. 2. Put the base checkpoint in `ComfyUI/models/diffusion_models/` - e.g. `qwen_image_2512_fp8_e4m3fn.safetensors` (20.43 GB, the file the official 2512 template names) or `qwen_image_2512_bf16.safetensors` (40.86 GB), from huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/split_files/diffusion_models. Edit checkpoints come from Comfy-Org/Qwen-Image-Edit_ComfyUI. Load with `Load Diffusion Model` (UNETLoader), not a checkpoint loader - these are unet-only files. 3. Put `qwen_2.5_vl_7b_fp8_scaled.safetensors` (9.38 GB) in `ComfyUI/models/text_encoders/` and load with `Load CLIP` (CLIPLoader), type `qwen_image`. Every official template uses exactly this pair. 4. Put `qwen_image_vae.safetensors` in `ComfyUI/models/vae/` and load with `Load VAE`. 5. Put LoRA `.safetensors` in `ComfyUI/models/loras/`. Subfolders show up as `subfolder/name.safetensors` in the dropdown - useful for separating t2i / edit / accelerator LoRAs. 6. Load with `LoraLoaderModelOnly`, patching only the MODEL line between the UNETLoader and ModelSamplingAuraFlow. Every official Comfy-Org Qwen template uses LoraLoaderModelOnly; the two-output `LoraLoader` also works but its CLIP branch is a no-op for DiT-only LoRAs. 7. Insert `ModelSamplingAuraFlow` after the LoRA chain. Use 3.1 for T2I and Edit-2511, 3.0 for the 2-step Turbo and Edit-2509 templates. Without this node ComfyUI uses the model's built-in shift of 1.15 and results differ from every published example. 8. Set KSampler to euler / simple / denoise 1.0, then steps+cfg per the accelerator (50/4.0 plain, 8/1.0 or 4/1.0 with Lightning, 2/1.0 with the 2-step Turbo). Latent from `EmptySD3LatentImage` at an official resolution. 9. Low-VRAM alternative: the `ComfyUI-GGUF` custom node with `Unet Loader (GGUF)` and a Qwen-Image/2512 GGUF (unsloth/Qwen-Image-2512-GGUF). LoRA loading still goes through LoraLoaderModelOnly, placed after the GGUF loader. 10. For Edit workflows, swap CLIPTextEncode for `TextEncodeQwenImageEdit` (single image) or `TextEncodeQwenImageEditPlus` (2509/2511) and add a `CFGNorm` node as the Edit-2511 template does; the LoRA chain sits in the same place. ─────────────────────────────────────────────────────────────────────────────── Scraped by lorakit on 2026-09-05 00:56 UTC. Stats and description are the author's; pros/cons are derived from the evidence tagged beside each line. ───────────────────────────────────────────────────────────────────────────────