Mirror published Civitai release: Storyflow Expression Card
Browse files
storyflow/expression-card/README.md
ADDED
|
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Storyflow Expression Card
|
| 2 |
+
|
| 3 |
+
Published by **Coffmeteor** for **Krea 2**.
|
| 4 |
+
|
| 5 |
+
[Original Civitai release](https://civitai.red/models/2906931?modelVersionId=3287591) · Version: v1.0
|
| 6 |
+
|
| 7 |
+
The files below preserve the exact published Civitai bytes. SHA-256 checksums are included.
|
| 8 |
+
|
| 9 |
+
## Downloads
|
| 10 |
+
|
| 11 |
+
| File | Type |
|
| 12 |
+
|---|---|
|
| 13 |
+
| [storyflow-expression-card-v2-lokr.safetensors](storyflow-expression-card-v2-lokr.safetensors) | Model |
|
| 14 |
+
|
| 15 |
+
## Usage
|
| 16 |
+
|
| 17 |
+
The Civitai trigger field is empty; follow the published usage notes below.
|
| 18 |
+
|
| 19 |
+
Verified default adapter strength: 1.
|
| 20 |
+
|
| 21 |
+
- SHA-256bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db
|
| 22 |
+
|
| 23 |
+
- Steps8
|
| 24 |
+
|
| 25 |
+
- CFG1.0
|
| 26 |
+
|
| 27 |
+
- SamplerEuler
|
| 28 |
+
|
| 29 |
+
- SchedulerBeta
|
| 30 |
+
|
| 31 |
+
- Denoise1.0
|
| 32 |
+
|
| 33 |
+
- Text encoderqwen3vl_4b_fp8_scaled.safetensors
|
| 34 |
+
|
| 35 |
+
- VAEqwen/qwen_image_vae.safetensors
|
| 36 |
+
|
| 37 |
+
[CHARACTER]
|
| 38 |
+
|
| 39 |
+
[CARD]
|
| 40 |
+
|
| 41 |
+
expression
|
| 42 |
+
|
| 43 |
+
[LAYOUT]
|
| 44 |
+
|
| 45 |
+
2x2 equal portrait grid, read left to right and top to bottom
|
| 46 |
+
|
| 47 |
+
[PANELS]
|
| 48 |
+
|
| 49 |
+
top-left normal: relaxed neutral face, closed mouth, calm eyes
|
| 50 |
+
|
| 51 |
+
top-right happy: genuine broad smile, visible upper teeth, raised cheeks
|
| 52 |
+
|
| 53 |
+
bottom-left angry: lowered knitted brows, hard stare, pressed lips, tense jaw
|
| 54 |
+
|
| 55 |
+
bottom-right surprised: raised eyebrows, wide-open eyes, rounded open mouth
|
| 56 |
+
|
| 57 |
+
[WARDROBE]
|
| 58 |
+
|
| 59 |
+
[RENDER]
|
| 60 |
+
|
| 61 |
+
realistic studio photograph; front-facing; neutral light-gray background; identical lighting and consistent chest-up framing in all four panels; no text, no labels, no captions
|
| 62 |
+
|
| 63 |
+
### Published model notes
|
| 64 |
+
|
| 65 |
+
- SHA-256bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db
|
| 66 |
+
|
| 67 |
+
- Steps8
|
| 68 |
+
|
| 69 |
+
- CFG1.0
|
| 70 |
+
|
| 71 |
+
- SamplerEuler
|
| 72 |
+
|
| 73 |
+
- SchedulerBeta
|
| 74 |
+
|
| 75 |
+
- Denoise1.0
|
| 76 |
+
|
| 77 |
+
- Text encoderqwen3vl_4b_fp8_scaled.safetensors
|
| 78 |
+
|
| 79 |
+
- VAEqwen/qwen_image_vae.safetensors
|
| 80 |
+
|
| 81 |
+
[CHARACTER]
|
| 82 |
+
|
| 83 |
+
[CARD]
|
| 84 |
+
|
| 85 |
+
expression
|
| 86 |
+
|
| 87 |
+
[LAYOUT]
|
| 88 |
+
|
| 89 |
+
2x2 equal portrait grid, read left to right and top to bottom
|
| 90 |
+
|
| 91 |
+
[PANELS]
|
| 92 |
+
|
| 93 |
+
top-left normal: relaxed neutral face, closed mouth, calm eyes
|
| 94 |
+
|
| 95 |
+
top-right happy: genuine broad smile, visible upper teeth, raised cheeks
|
| 96 |
+
|
| 97 |
+
bottom-left angry: lowered knitted brows, hard stare, pressed lips, tense jaw
|
| 98 |
+
|
| 99 |
+
bottom-right surprised: raised eyebrows, wide-open eyes, rounded open mouth
|
| 100 |
+
|
| 101 |
+
[WARDROBE]
|
| 102 |
+
|
| 103 |
+
[RENDER]
|
| 104 |
+
|
| 105 |
+
realistic studio photograph; front-facing; neutral light-gray background; identical lighting and consistent chest-up framing in all four panels; no text, no labels, no captions
|
| 106 |
+
|
| 107 |
+
## Provenance and terms
|
| 108 |
+
|
| 109 |
+
The original public descriptions, exact trigger field, permission flags and file metadata are retained in [source.json](source.json). Original model and component terms remain applicable; this mirror does not grant a different license.
|
| 110 |
+
|
| 111 |
+
[SHA-256 checksums](SHA256SUMS)
|
storyflow/expression-card/SHA256SUMS
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db storyflow-expression-card-v2-lokr.safetensors
|
storyflow/expression-card/source.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_id": 2906931,
|
| 3 |
+
"version_id": 3287591,
|
| 4 |
+
"title": "Storyflow Expression Card",
|
| 5 |
+
"source_model_name": "Storyflow Expression Card",
|
| 6 |
+
"source_version_name": "v1.0",
|
| 7 |
+
"base_model": "Krea 2",
|
| 8 |
+
"type": "LORA",
|
| 9 |
+
"source_url": "https://civitai.red/models/2906931?modelVersionId=3287591",
|
| 10 |
+
"trained_words": [],
|
| 11 |
+
"verified_usage": [
|
| 12 |
+
{
|
| 13 |
+
"trigger": "",
|
| 14 |
+
"default_strength": 1
|
| 15 |
+
}
|
| 16 |
+
],
|
| 17 |
+
"source_model_description_html": "<p>备注:平时自己做视频会用的,仅供参考,大部分模型本身就可以做到这些,LoRA只是提高了一下效率。</p><p>一、简介</p><p>Storyflow Expression Card 用于一次生成同一人物的 2×2 表情卡,并固定为以下顺序:</p><ol><li><p>左上:正常;</p></li><li><p>右上:开心;</p></li><li><p>左下:生气;</p></li><li><p>右下:惊讶。</p></li></ol><p>它适合角色设定、表情参考、故事板角色资产和后续角色一致性工作流。文件采用 LoKr 结构,在 ComfyUI 中仍通过普通 <code>Load LoRALoraLoader</code> 节点加载。</p><p>二、模型文件</p><p>- 文件名<code>storyflow-expression-card-v2-lokr.safetensors</code></p><p>- SHA-256<code>bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db</code></p><p>- 推荐强度:Model <code>0.7</code>、CLIP <code>0.7</code></p><p>- 已测试底模:Krea 2 Turbo BF16、Moody-Krea BF16</p><p>三、推荐参数</p><p>- 分辨率<code>1024 × 1280</code></p><p>- Steps<code>8</code></p><p>- CFG<code>1.0</code></p><p>- Sampler<code>Euler</code></p><p>- Scheduler<code>Beta</code></p><p>- Denoise<code>1.0</code></p><p>- Text encoder<code>qwen3vl_4b_fp8_scaled.safetensors</code></p><p>- VAE<code>qwen/qwen_image_vae.safetensors</code></p><p>四、使用方法</p><p>1. 将模型放入 <code>ComfyUI/models/loras/krea2/</code>。</p><p>2. 在底模和文本编码器之后连接 <code>Load LoRA</code>,Model/CLIP 强度均设为 <code>0.7</code>。</p><p>3. 使用下面的 Prompt 结构,保持四个槽位的名称与顺序不变。</p><p>4. 按推荐参数生成。也可以直接把本目录中的 <code>workflow-*.png</code> 拖进 ComfyUI,读取完整节点和参数。</p><p>五、Prompt 模板</p><pre><code>[CHARACTER]\n\n<人物身份、年龄、面部、肤色、发型等描述>; exactly the same person in every panel\n\n[CARD]\n\nexpression\n\n[LAYOUT]\n\n2x2 equal portrait grid, read left to right and top to bottom\n\n[PANELS]\n\ntop-left normal: relaxed neutral face, closed mouth, calm eyes\n\ntop-right happy: genuine broad smile, visible upper teeth, raised cheeks\n\nbottom-left angry: lowered knitted brows, hard stare, pressed lips, tense jaw\n\nbottom-right surprised: raised eyebrows, wide-open eyes, rounded open mouth\n\n[WARDROBE]\n\n<服装描述> in every panel\n\n[RENDER]\n\nrealistic studio photograph; front-facing; neutral light-gray background; identical lighting and consistent chest-up framing in all four panels; no text, no labels, no captions</code></pre><p>建议把人物身份和服装写得具体,但不要改变 <code>[PANELS]</code> 的顺序。若需要非写实风格,可修改 <code>[RENDER]</code>,同时保留固定网格和一致性约束。</p><p>六、发布图说明</p><p>- <code>0102</code>:分别在 Turbo BF16 和 Moody-Krea 下进行同 prompt、同 seed 的 LoRA OFF/ON 0.7 全览对比。</p><p>- <code>0306</code>:LoRA ON 0.7 原始 PNG,均内含 ComfyUI <code>prompt</code> 与 <code>workflow</code> 元数据。</p>",
|
| 18 |
+
"source_version_description_html": "<p>一、简介</p><p>Storyflow Expression Card 用于一次生成同一人物的 2×2 表情卡,并固定为以下顺序:</p><ol><li><p>左上:正常;</p></li><li><p>右上:开心;</p></li><li><p>左下:生气;</p></li><li><p>右下:惊讶。</p></li></ol><p>它适合角色设定、表情参考、故事板角色资产和后续角色一致性工作流。文件采用 LoKr 结构,在 ComfyUI 中仍通过普通 <code>Load LoRALoraLoader</code> 节点加载。</p><p>二、模型文件</p><p>- 文件名<code>storyflow-expression-card-v2-lokr.safetensors</code></p><p>- SHA-256<code>bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db</code></p><p>- 推荐强度:Model <code>0.7</code>、CLIP <code>0.7</code></p><p>- 已测试底模:Krea 2 Turbo BF16、Moody-Krea BF16</p><p>三、推荐参数</p><p>- 分辨率<code>1024 × 1280</code></p><p>- Steps<code>8</code></p><p>- CFG<code>1.0</code></p><p>- Sampler<code>Euler</code></p><p>- Scheduler<code>Beta</code></p><p>- Denoise<code>1.0</code></p><p>- Text encoder<code>qwen3vl_4b_fp8_scaled.safetensors</code></p><p>- VAE<code>qwen/qwen_image_vae.safetensors</code></p><p>四、使用方法</p><p>1. 将模型放入 <code>ComfyUI/models/loras/krea2/</code>。</p><p>2. 在底模和文本编码器之后连接 <code>Load LoRA</code>,Model/CLIP 强度均设为 <code>0.7</code>。</p><p>3. 使用下面的 Prompt 结构,保持四个槽位的名称与顺序不变。</p><p>4. 按推荐参数生成。也可以直接把本目录中的 <code>workflow-*.png</code> 拖进 ComfyUI,读取完整节点和参数。</p><p>五、Prompt 模板</p><pre><code>[CHARACTER]\n\n<人物身份、年龄、面部、肤色、发型等描述>; exactly the same person in every panel\n\n[CARD]\n\nexpression\n\n[LAYOUT]\n\n2x2 equal portrait grid, read left to right and top to bottom\n\n[PANELS]\n\ntop-left normal: relaxed neutral face, closed mouth, calm eyes\n\ntop-right happy: genuine broad smile, visible upper teeth, raised cheeks\n\nbottom-left angry: lowered knitted brows, hard stare, pressed lips, tense jaw\n\nbottom-right surprised: raised eyebrows, wide-open eyes, rounded open mouth\n\n[WARDROBE]\n\n<服装描述> in every panel\n\n[RENDER]\n\nrealistic studio photograph; front-facing; neutral light-gray background; identical lighting and consistent chest-up framing in all four panels; no text, no labels, no captions</code></pre><p>建议把人物身份和服装写得具体,但不要改变 <code>[PANELS]</code> 的顺序。若需要非写实风格,可修改 <code>[RENDER]</code>,同时保留固定网格和一致性约束。</p><p>六、发布图说明</p><p>- <code>0102</code>:分别在 Turbo BF16 和 Moody-Krea 下进行同 prompt、同 seed 的 LoRA OFF/ON 0.7 全览对比。</p><p>- <code>0306</code>:LoRA ON 0.7 原始 PNG,均内含 ComfyUI <code>prompt</code> 与 <code>workflow</code> 元数据。</p>",
|
| 19 |
+
"permissions": {
|
| 20 |
+
"allowNoCredit": true,
|
| 21 |
+
"allowCommercialUse": [
|
| 22 |
+
"Image",
|
| 23 |
+
"RentCivit",
|
| 24 |
+
"Rent",
|
| 25 |
+
"Sell",
|
| 26 |
+
"SellMerge"
|
| 27 |
+
],
|
| 28 |
+
"allowDerivatives": true,
|
| 29 |
+
"allowDifferentLicense": true
|
| 30 |
+
},
|
| 31 |
+
"files": [
|
| 32 |
+
{
|
| 33 |
+
"id": 3171971,
|
| 34 |
+
"name": "storyflow-expression-card-v2-lokr.safetensors",
|
| 35 |
+
"type": "Model",
|
| 36 |
+
"path": "storyflow/expression-card/storyflow-expression-card-v2-lokr.safetensors",
|
| 37 |
+
"sha256": "bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db",
|
| 38 |
+
"size_kb": 57388.4609375,
|
| 39 |
+
"metadata": {
|
| 40 |
+
"format": "SafeTensor",
|
| 41 |
+
"fp": "bf16",
|
| 42 |
+
"isRequired": false
|
| 43 |
+
},
|
| 44 |
+
"download_url": "https://civitai.red/api/download/models/3287591?fileId=3171971"
|
| 45 |
+
}
|
| 46 |
+
]
|
| 47 |
+
}
|
storyflow/expression-card/storyflow-expression-card-v2-lokr.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bdcec9c8cea41853264c9809fd9bd37be80aa36038d7d5a971d26f0c2c1698db
|
| 3 |
+
size 58765784
|