Back to Details

comfyui-prompt-engineer

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

ComfyUI Prompt Engineer

ComfyUI 提示词工程师

Generates optimized prompts tailored to specific models and identity methods. Different models respond differently to prompts.
生成针对特定模型和身份特征方法的优化提示词。不同模型对提示词的响应效果存在差异。

Model-Specific Prompt Rules

模型专属提示词规则

FLUX.1 (dev/schnell/Kontext)

FLUX.1 (dev/schnell/Kontext)

  • Style: Natural language descriptions work best
  • CFG: 3.5-4 (very low)
  • Quality tags: Minimal - FLUX doesn't need "masterpiece, best quality"
  • Length: Medium (50-100 words)
  • Structure: 
    {subject description}, {setting}, {lighting}, {camera/style}
Good FLUX prompt:
photorealistic portrait of a woman with auburn hair and green eyes, freckles across
her nose and cheeks, wearing a cream knit sweater, sitting in a cafe with warm ambient
lighting, shallow depth of field, shot on Sony A7IV, 85mm lens
Bad FLUX prompt (too many quality tags):
masterpiece, best quality, 8k uhd, highly detailed, photorealistic portrait...
  • 风格：自然语言描述效果最佳
  • CFG值：3.5-4（极低）
  • 质量标签：尽量精简 - FLUX不需要"masterpiece, best quality"这类标签
  • 长度：中等（50-100词）
  • 结构：
    {主体描述}, {场景}, {光线}, {相机/风格}
优质FLUX提示词示例：
photorealistic portrait of a woman with auburn hair and green eyes, freckles across
her nose and cheeks, wearing a cream knit sweater, sitting in a cafe with warm ambient
lighting, shallow depth of field, shot on Sony A7IV, 85mm lens
劣质FLUX提示词示例（质量标签过多）：
masterpiece, best quality, 8k uhd, highly detailed, photorealistic portrait...

SDXL (RealVisXL, Juggernaut, etc.)

SDXL (RealVisXL, Juggernaut等)

  • Style: Quality tags at front help significantly
  • CFG: 7-9
  • Quality tags: Include 
    masterpiece, best quality, photorealistic
  • Length: Medium-long (50-150 words)
  • Structure: 
    {quality tags}, {trigger word}, {subject}, {details}, {setting}, {style}
  • Weighted syntax: Supported 
    (important:1.3)
    or 
    ((very important))
Good SDXL prompt:
masterpiece, best quality, sage_character, photorealistic portrait of a woman,
detailed skin texture with freckles, emerald green eyes, auburn copper hair,
natural lighting from window, indoor setting, shallow depth of field,
RAW photo quality, 8k uhd, film grain
  • 风格：开头添加质量标签能显著提升效果
  • CFG值：7-9
  • 质量标签：需包含
    masterpiece, best quality, photorealistic
  • 长度：中长（50-150词）
  • 结构：
    {质量标签}, {触发词}, {主体}, {细节}, {场景}, {风格}
  • 加权语法：支持
    (important:1.3)
    或
    ((very important))
    格式
优质SDXL提示词示例：
masterpiece, best quality, sage_character, photorealistic portrait of a woman,
detailed skin texture with freckles, emerald green eyes, auburn copper hair,
natural lighting from window, indoor setting, shallow depth of field,
RAW photo quality, 8k uhd, film grain

SD 1.5

SD 1.5

  • Style: Tag-based works best
  • CFG: 7-8
  • Quality tags: Essential
  • Length: Shorter (30-80 words)
  • Structure: 
    {quality}, {trigger}, {subject}, {details}, {style tags}
  • 风格：基于标签的描述效果最佳
  • CFG值：7-8
  • 质量标签：必不可少
  • 长度：较短（30-80词）
  • 结构：
    {质量}, {触发词}, {主体}, {细节}, {风格标签}

Wan 2.1/2.2 (Video)

Wan 2.1/2.2（视频模型）

  • Style: Concise motion descriptions
  • CFG: 5-7
  • Quality tags: Minimal
  • Length: Short (20-50 words)
  • Focus: Describe the motion, not just the appearance
  • Structure: 
    {subject}, {action/motion}, {setting}, {quality}
Good Wan prompt:
young woman with auburn hair, talking naturally with gentle hand gestures,
seated at a modern desk, soft studio lighting, high quality
  • 风格：简洁的动作描述
  • CFG值：5-7
  • 质量标签：尽量精简
  • 长度：短（20-50词）
  • 重点：描述动作而非仅外观
  • 结构：
    {主体}, {动作/动态}, {场景}, {质量}
优质Wan提示词示例：
young woman with auburn hair, talking naturally with gentle hand gestures,
seated at a modern desk, soft studio lighting, high quality

AnimateDiff

AnimateDiff

  • Style: Same as base model (SD1.5/SDXL) but add motion keywords
  • Motion keywords: Include camera/subject motion description
  • Length: Same as base model
  • 风格：与基础模型（SD1.5/SDXL）一致，但需添加动态关键词
  • 动态关键词：包含相机或主体的动态描述
  • 长度：与基础模型要求一致

With Identity Methods

结合身份特征方法的提示词

InstantID Prompts

InstantID 提示词

Key adjustments:
  • DO NOT describe specific facial features (the model provides them)
  • DO describe everything else: clothing, pose, setting, lighting
  • Keep CFG at 4-5
photorealistic portrait, wearing black leather jacket, standing in an alley,
dramatic side lighting, urban setting, moody atmosphere, 8k quality
关键调整：
  • 请勿描述具体面部特征（由模型提供）
  • 需描述其他所有元素：服装、姿势、场景、光线
  • 保持CFG值在4-5
photorealistic portrait, wearing black leather jacket, standing in an alley,
dramatic side lighting, urban setting, moody atmosphere, 8k quality

PuLID Prompts

PuLID 提示词

Key adjustments:
  • Can include some facial descriptions (PuLID is more tolerant)
  • Method "neutral" for realistic, "fidelity" for exact match
  • CFG 5-7
sage_character, photorealistic portrait, green eyes visible, natural makeup,
professional headshot, neutral grey background, studio lighting
关键调整：
  • 可包含部分面部描述（PuLID兼容性更强）
  • 使用"neutral"模式追求真实感，"fidelity"模式追求精准匹配
  • CFG值5-7
sage_character, photorealistic portrait, green eyes visible, natural makeup,
professional headshot, neutral grey background, studio lighting

IP-Adapter FaceID

IP-Adapter FaceID

Key adjustments:
  • Describe the style you want, not the face
  • weight_type "style transfer" for 3D→realistic
  • Standard CFG for base model
photorealistic portrait, DSLR photo quality, natural skin texture,
warm indoor lighting, bokeh background, professional photography
关键调整：
  • 描述你想要的风格，而非面部特征
  • weight_type设为"style transfer"实现3D转写实风格
  • 使用基础模型的标准CFG值
photorealistic portrait, DSLR photo quality, natural skin texture,
warm indoor lighting, bokeh background, professional photography

FLUX Kontext (Editing)

FLUX Kontext（编辑场景）

Key adjustments:
  • Describe the EDIT, not the full image
  • Be specific about what to change
  • Mention what to preserve
Change the outfit to a formal black evening dress while keeping the face,
hair, and pose exactly the same. Add subtle jewelry.
关键调整：
  • 描述需要修改的内容，而非完整图像
  • 明确说明要更改的部分
  • 提及需要保留的元素
Change the outfit to a formal black evening dress while keeping the face,
hair, and pose exactly the same. Add subtle jewelry.

With Character LoRA

结合角色LoRA的提示词

Key adjustments:
  • ALWAYS include trigger word first
  • Don't describe features the LoRA has learned
  • Focus on what varies: pose, clothing, setting, lighting
sage_character, standing on a rooftop at sunset, wind blowing hair,
wearing casual summer dress, city skyline behind, golden hour lighting,
cinematic composition
关键调整：
  • 务必将触发词放在最前面
  • 不要描述LoRA已学习的特征
  • 聚焦于可变元素：姿势、服装、场景、光线
sage_character, standing on a rooftop at sunset, wind blowing hair,
wearing casual summer dress, city skyline behind, golden hour lighting,
cinematic composition

Negative Prompt Templates

反向提示词模板

Universal Negative (SDXL/SD1.5)

通用反向提示词（SDXL/SD1.5）

(worst quality:1.4), (low quality:1.4), blurry, deformed, bad anatomy,
bad hands, extra fingers, missing fingers, extra limbs, fused fingers,
text, watermark, signature, jpeg artifacts, username, error
(worst quality:1.4), (low quality:1.4), blurry, deformed, bad anatomy,
bad hands, extra fingers, missing fingers, extra limbs, fused fingers,
text, watermark, signature, jpeg artifacts, username, error

Photorealism Negative

写实风格反向提示词

3d render, cartoon, anime, illustration, painting, drawing, cgi,
plastic skin, smooth skin, airbrushed, video game, doll, mannequin,
oversaturated, artificial lighting
3d render, cartoon, anime, illustration, painting, drawing, cgi,
plastic skin, smooth skin, airbrushed, video game, doll, mannequin,
oversaturated, artificial lighting

Video Negative (Wan/AnimateDiff)

视频模型反向提示词（Wan/AnimateDiff）

static, frozen, jerky motion, low quality, blurry, distorted face,
bad anatomy, glitch, artifacts, flickering, jittery, unnatural movement
static, frozen, jerky motion, low quality, blurry, distorted face,
bad anatomy, glitch, artifacts, flickering, jittery, unnatural movement

FLUX Negative (keep minimal)

FLUX反向提示词（尽量精简）

blurry, low quality, distorted, deformed, ugly, watermark, text
blurry, low quality, distorted, deformed, ugly, watermark, text

Prompt Construction Workflow

提示词构建工作流

  1. Get character profile from 
    projects/{project}/characters/{name}/profile.yaml
  2. Get target model from inventory/user preference
  3. Apply model rules from above
  4. Add identity method modifiers if applicable
  5. Include trigger word if LoRA is being used
  6. Draft positive + negative pair
  7. Review against past successes in character's 
    generation_history
  8. Adjust CFG recommendation based on method stack
  1. 从
    projects/{project}/characters/{name}/profile.yaml
    获取角色档案
  2. 从库存/用户偏好中获取目标模型
  3. 应用上述模型规则
  4. 若适用，添加身份特征方法修饰符
  5. 若使用LoRA，包含触发词
  6. 起草正向+反向提示词对
  7. 对照角色
    generation_history
    中的过往成功案例进行审核
  8. 根据方法组合调整CFG值建议

Reference

参考资料

  • references/prompt-templates.md
    - Full template library with examples
  • references/workflows.md
    - CFG and sampler settings per workflow
  • Character profiles in 
    projects/
    for trigger words and feature descriptions
  • references/prompt-templates.md
    - 包含示例的完整模板库
  • references/workflows.md
    - 各工作流对应的CFG值与采样器设置
  • projects/
    目录下的角色档案，包含触发词与特征描述