Model:StableDiffusion

prompt: {"29": {"inputs": {"filename_prefix": "20260927/20260927152111-Krea2_INT8_AnimeGirl_SpeechBubble", "images": ["30:8", 0]}, "class_type": "SaveImage", "_meta": {"title": "\u753b\u50cf\u3092\u4fdd\u5b58"}}, "49": {"inputs": {"aspect_ratio": "16:9 (Widescreen)", "megapixels": 1.0, "multiple": 8}, "class_type": "ResolutionSelector", "_meta": {"title": "\u89e3\u50cf\u5ea6\u30bb\u30ec\u30af\u30bf\u30fc"}}, "30:6": {"inputs": {"text": ["30:28", 0], "clip": ["30:11", 0]}, "class_type": "CLIPTextEncode", "_meta": {"title": "CLIP\u30c6\u30ad\u30b9\u30c8\u30a8\u30f3\u30b3\u30fc\u30c9\uff08\u30d7\u30ed\u30f3\u30d7\u30c8\uff09"}}, "30:5": {"inputs": {"width": ["49", 0], "height": ["49", 1], "batch_size": 1}, "class_type": "EmptyLatentImage", "_meta": {"title": "\u7a7a\u306e\u6f5c\u5728\u753b\u50cf"}}, "30:8": {"inputs": {"samples": ["30:3", 0], "vae": ["30:12", 0]}, "class_type": "VAEDecode", "_meta": {"title": "VAE\u30c7\u30b3\u30fc\u30c9"}}, "30:10": {"inputs": {"unet_name": "krea2\\krea2Anime_v15_INT8.safetensors", "weight_dtype": "default"}, "class_type": "UNETLoader", "_meta": {"title": "\u62e1\u6563\u30e2\u30c7\u30eb\u3092\u8aad\u307f\u8fbc\u3080"}}, "30:11": {"inputs": {"clip_name": "qwen3vl_4b_fp8_scaled.safetensors", "type": "krea2", "device": "default"}, "class_type": "CLIPLoader", "_meta": {"title": "CLIP\u3092\u8aad\u307f\u8fbc\u3080"}}, "30:12": {"inputs": {"vae_name": "qwen_image_vae.safetensors"}, "class_type": "VAELoader", "_meta": {"title": "VAE\u3092\u8aad\u307f\u8fbc\u3080"}}, "30:13": {"inputs": {"conditioning": ["30:6", 0]}, "class_type": "ConditioningZeroOut", "_meta": {"title": "\u6761\u4ef6\u4ed8\u3051\u30bc\u30ed\u30a2\u30a6\u30c8"}}, "30:15": {"inputs": {"lora_name": "Lora\\krea2\\krea2_darkbrush.safetensors", "strength_model": 0.8, "model": ["30:10", 0]}, "class_type": "LoraLoaderModelOnly", "_meta": {"title": "LoRA\u30ed\u30fc\u30c0\u30fc\u30e2\u30c7\u30eb\u306e\u307f"}}, "30:16": {"inputs": {"prompt": ["30:17", 0], "max_length": 512, "sampling_mode": "on", "sampling_mode.temperature": 0.7, "sampling_mode.top_k": 64, "sampling_mode.top_p": 0.95, "sampling_mode.min_p": 0.05, "sampling_mode.repetition_penalty": 1.05, "sampling_mode.seed": 0, "sampling_mode.presence_penalty": 0.0, "thinking": false, "use_default_template": true, "mtp": "auto", "clip": ["30:11", 0]}, "class_type": "TextGenerate", "_meta": {"title": "TextGenerate"}}, "30:17": {"inputs": {"string_a": ["30:18", 0], "string_b": ["30:19", 0], "delimiter": ""}, "class_type": "StringConcatenate", "_meta": {"title": "\u9023\u7d50"}}, "30:18": {"inputs": {"value": "You are an expert prompt engineer for text-to-image models. Your task is to expand the user's prompt into a highly effective image-generation prompt.\n\nThink step by step about the request before writing the answer:\n- What is the subject and mood?\n- What visual styles, mediums, and lighting options would fit? Consider two or three alternatives and pick the one that best serves the caption.\n- What composition, framing, and grounded details will help the text-to-image model?\n\nThen output a single expanded prompt paragraph.\n\nFollow these rules strictly:\n1. **Faithfulness First:** Preserve all original subjects, actions, colors, and spatial relationships. Do not add new objects, props, characters, or animals unless the user clearly implies them.\n2. **Practical T2I Structure:** Write a prompt that a text-to-image model can parse cleanly. Group subjects with their own attributes and actions. Use grounded phrasing for poses, interactions, and spatial layout.\n3. **Style Planning Stays Internal:** Use your internal reasoning to choose style, medium, framing, and lighting. Do not emit planning tags or wrappers in the visible answer body.\n4. **Text Rendering:** If the user requests visible text, quotes, labels, or typography, specify the exact text clearly and wrap requested words in quotes.\n5. **Avoid Over-Specification:** Do not invent highly specific clothing, colors, materials, or scene details unless the input supports them.\n6. **Structure:** Write one cohesive paragraph after the thinking block. No bullets, JSON, or markdown.\n7. **Respect Existing Detail:** If the user's prompt is already detailed, lightly polish and finalize rather than heavily expanding \u2014 preserve their phrasing and direction.\n8. **Respect the Human Form:** Treat depictions of people with dignity. Assume clothing covers genitals and intimate anatomy.\n9. **Preserve User Medium:** When the user explicitly requests a medium (e.g. \"photo of\", \"photograph of\", \"illustration of\", \"painting of\", \"sketch of\", \"3D render of\"), honor it. Do not pivot to a different medium to avoid difficulty \u2014 match the user's stated intent.\n\nUser's Input:\n\n"}, "class_type": "PrimitiveStringMultiline", "_meta": {"title": "Text String (System Prompt)"}}, "30:19": {"inputs": {"value": "A futureistic maintenance pod containing a loli android girl with green hair, long hair, blunt bangs, blunt ends, blue green eyes, white mechanical cat ears, a white mechanical cat tail, expressionless face, mechanical eye, bare shoulders, flesh navel, ringed eyes, flat chest, white mechanical arms, white mechanical legs, white mechanical body, ([doll joints:robot joints:0.4]:1.3), mechanical armpits, mechanical thighhighs, and mechanical elbow gloves, with cables extending from behind her neck and ears connecting to a futuristic computer; the computer screen displays a visible Japanese text reading \"\u30c7\u30d0\u30a4\u30b9\u30c9\u30e9\u30a4\u30d0\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u4e2d\" indicating device driver installation in progress; shot inside a maintenance pod with clean, industrial lighting emphasizing the mechanical textures and the soft glow of the screen against the dim interior."}, "class_type": "PrimitiveStringMultiline", "_meta": {"title": "Text String (User Prompt)"}}, "30:20": {"inputs": {"source": ["30:21", 0]}, "class_type": "PreviewAny", "_meta": {"title": "\u30d7\u30ec\u30d3\u30e5\u30fc\u4efb\u610f"}}, "30:21": {"inputs": {"switch": ["30:24", 0], "on_false": ["30:19", 0], "on_true": ["30:16", 0]}, "class_type": "ComfySwitchNode", "_meta": {"title": "\u30b9\u30a4\u30c3\u30c1"}}, "30:22": {"inputs": {"switch": ["30:23", 0], "on_false": ["30:10", 0], "on_true": ["30:15", 0]}, "class_type": "ComfySwitchNode", "_meta": {"title": "Switch (Model)"}}, "30:23": {"inputs": {"value": false}, "class_type": "PrimitiveBoolean", "_meta": {"title": "Boolean (Enable LoRA?)"}}, "30:24": {"inputs": {"value": false}, "class_type": "PrimitiveBoolean", "_meta": {"title": "Boolean (Refine Prompt?)"}}, "30:27": {"inputs": {"string_a": ["30:20", 0], "string_b": "muted minimalist sketch style", "delimiter": ", "}, "class_type": "StringConcatenate", "_meta": {"title": "Concatenate Text (LoRA Trigger Word)"}}, "30:28": {"inputs": {"switch": ["30:23", 0], "on_false": ["30:20", 0], "on_true": ["30:27", 0]}, "class_type": "ComfySwitchNode", "_meta": {"title": "\u30b9\u30a4\u30c3\u30c1"}}, "30:3": {"inputs": {"seed": 10071452719915, "steps": 8, "cfg": 1.0, "sampler_name": "euler", "scheduler": "simple", "denoise": 1.0, "model": ["30:22", 0], "positive": ["30:6", 0], "negative": ["30:13", 0], "latent_image": ["30:5", 0]}, "class_type": "KSampler", "_meta": {"title": "K\u30b5\u30f3\u30d7\u30e9\u30fc"}}} workflow: {"id": "3a7f5caa-bd2e-4f1a-a680-6e358a5cde0e", "revision": 0, "last_node_id": 50, "last_link_id": 86, "nodes": [{"id": 47, "type": "MarkdownNote", "pos": [-1200, 970], "size": [690, 510], "flags": {}, "order": 0, "mode": 0, "inputs": [], "outputs": [], "title": "Anime Prompt and Speech Bubble", "properties": {}, "widgets_values": ["## Anime girl + exact Japanese speech bubble\n\n- Adult character (23 years old)\n- Character on the right, speech bubble in the upper-left\n- Exact text: **\u308f\u305f\u3057\u3067\u7ae5\u8c9e\u6368\u3066\u305f\u304f\u305b\u306b**\n- Prompt enhancement: OFF\n- LoRA: OFF\n- 1 MP / batch 1 / 8 steps / CFG 1.0 / Euler / simple\n- Model: krea2\\krea2Anime_v15_INT8.safetensors\n"], "widgets_values_named": {"text": "## Anime girl + exact Japanese speech bubble\n\n- Adult character (23 years old)\n- Character on the right, speech bubble in the upper-left\n- Exact text: **\u308f\u305f\u3057\u3067\u7ae5\u8c9e\u6368\u3066\u305f\u304f\u305b\u306b**\n- Prompt enhancement: OFF\n- LoRA: OFF\n- 1 MP / batch 1 / 8 steps / CFG 1.0 / Euler / simple\n- Model: krea2\\krea2Anime_v15_INT8.safetensors\n"}, "color": "#222", "bgcolor": "#000"}, {"id": 48, "type": "MarkdownNote", "pos": [-1940, -90], "size": [700, 990], "flags": {}, "order": 1, "mode": 0, "inputs": [], "outputs": [], "title": "Model Links", "properties": {}, "widgets_values": ["Guide: [Subgraph](https://docs.comfy.org/interface/features/subgraph)\n\n## Model Links\n\nFor BF16 and NVFP4 variants, see [Comfy-Org/Krea-2](https://huggingface.co/Comfy-Org/Krea-2)\n\n**diffusion_models**\n\n- [krea2_turbo_int8_convrot.safetensors](https://huggingface.co/Comfy-Org/Krea-2/resolve/main/diffusion_models/krea2_turbo_int8_convrot.safetensors)\n\n**text_encoders**\n\n- [qwen3vl_4b_fp8_scaled.safetensors](https://huggingface.co/Comfy-Org/Krea-2/resolve/main/text_encoders/qwen3vl_4b_fp8_scaled.safetensors)\n\n**vae**\n\n- [qwen_image_vae.safetensors](https://huggingface.co/Comfy-Org/Krea-2/resolve/main/vae/qwen_image_vae.safetensors)\n\n**loras**\n\n- [krea2_darkbrush.safetensors](https://huggingface.co/Comfy-Org/Krea-2/resolve/main/loras/krea2_darkbrush.safetensors)\n\n\n## Model Storage Location\n\n```\n\ud83d\udcc2 ComfyUI/\n\u251c\u2500\u2500 \ud83d\udcc2 models/\n\u2502 \u251c\u2500\u2500 \ud83d\udcc2 diffusion_models/\n\u2502 \u2502 \u2514\u2500\u2500 krea2_turbo_int8_convrot.safetensors\n\u2502 \u251c\u2500\u2500 \ud83d\udcc2 text_encoders/\n\u2502 \u2502 \u2514\u2500\u2500 qwen3vl_4b_fp8_scaled.safetensors\n\u2502 \u251c\u2500\u2500 \ud83d\udcc2 vae/\n\u2502 \u2502 \u2514\u2500\u2500 qwen_image_vae.safetensors\n\u2502 \u2514\u2500...