qwen-wanx-comic-gen

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Qwen Wanx 漫画生成器

Qwen Wanx Comic Generator

使用阿里云通义万相 wan2.6-t2i 文生图 API,将用户的文字描述转为漫画 / 动漫风格图片,适合单格插画和简单多格分镜草图。

Use Alibaba Cloud's Qwen Wanx wan2.6-t2i text-to-image API to convert users' text descriptions into comic/anime-style images, suitable for single-panel illustrations and simple multi-panel storyboard sketches.

何时使用本技能

When to Use This Skill

当用户明确提到：
- “用万相/通义万相/千问万相画图（画漫画）”
- “用 Qwen 生成漫画/二次元/动漫风格图片”
- “帮我生成一张漫画风格的插画”
当用户需要 偏漫画/动漫风格 的图像，而不是写实摄影风格时，可以优先考虑本技能。

When users explicitly mention:
- "Draw (comic) with Wanx/Qwen Wanx/Qwen Qianwen Wanx"
- "Generate comic/anime/ACG-style image with Qwen"
- "Help me generate a comic-style illustration"
When users need comic/anime-style images rather than realistic photography style, this skill can be prioritized.

运行方式

How to Run

在本地或远程主机上运行脚本（会自动复用 OpenClaw 主模型使用的 API Key）：

bash

python3 {baseDir}/scripts/gen.py \
  --prompt "黑白分镜,四格日常校园搞笑漫画,主角是一只拟人化的小龙虾" \
  --style anime \
  --size 1280*1280 \
  --n 1

更多示例:

bash

undefined

Run the script on local or remote host (it will automatically reuse the API Key used by the OpenClaw main model):

bash

python3 {baseDir}/scripts/gen.py \
  --prompt "黑白分镜,四格日常校园搞笑漫画,主角是一只拟人化的小龙虾" \
  --style anime \
  --size 1280*1280 \
  --n 1

More examples:

bash

undefined

生成彩色二次元人物立绘(--style 会自动添加"动漫风格"前缀)

Generate colored anime character full-body illustration (--style will automatically add "anime style" prefix)

python3 {baseDir}/scripts/gen.py --prompt "全身立绘,日系漫画风格,短发少女,背着双肩包" --style anime

生成黑白线稿分镜草图(--style sketch 会添加"素描风格"前缀)

Generate black-and-white lineart storyboard sketch (--style sketch will add "sketch style" prefix)

python3 {baseDir}/scripts/gen.py --prompt "四格漫画分镜,办公室日常吐槽,黑白线稿" --style sketch

使用反向提示词,避免写实风格

Use negative prompt to avoid realistic style

python3 {baseDir}/scripts/gen.py
--prompt "可爱的小龙虾超级英雄,日式漫画封面"
--negative-prompt "photography, realistic, photo, 3d"
--style anime

不使用风格前缀(手动在 prompt 里描述风格)

Do not use style prefix (describe style manually in prompt)

python3 {baseDir}/scripts/gen.py
--prompt "动漫风格,Q版校园日常,小龙虾吉祥物"
--style auto
--n 2

undefined

python3 {baseDir}/scripts/gen.py
--prompt "动漫风格,Q版校园日常,小龙虾吉祥物"
--style auto
--n 2

undefined

参数说明

Parameter Explanation

```
--prompt
```
(必选):中文或英文描述,重点写清:角色、场景、情绪、构图和"漫画/动漫风格"要求。

--negative-prompt

(可选):不想要的效果,如

photography, realistic, low quality

等。

```
--style
```
(可选):风格提示词(脚本会自动添加到 prompt 前面),默认
```
anime
```
,可选:
- ```
anime
```
  :添加"动漫风格,"前缀
- ```
flat illustration
```
  :添加"扁平插画风格,"前缀
- ```
3d cartoon
```
  :添加"3D卡通风格,"前缀
- ```
sketch
```
  :添加"素描风格,"前缀
- ```
auto
```
  :不添加风格前缀(手动在 prompt 里描述)
- 注意:wan2.6-t2i 不支持 style API 参数,脚本会把风格关键词融入 prompt。
```
--size
```
(可选):输出分辨率,默认
```
1280*1280
```
。wan2.6-t2i 支持:
- 总像素在 1280×1280 ~ 1440×1440 之间
- 宽高比在 1:4 ~ 4:1 之间
- 示例:
```
1280*1280
```
  (方图)、
```
768*2700
```
  (竖长图)、
```
1696*960
```
  (16:9 横图)
```
--n
```
(可选):一次生成的图片数量,1–4,默认 1。

--output-dir

(可选):输出目录,默认

./tmp/qwen-wanx-comic-<时间戳>

。

```
--prompt
```
(required): Chinese or English description, focus on clarifying: character, scene, emotion, composition, and "comic/anime style" requirements.

--negative-prompt

(optional): Unwanted effects, such as

photography, realistic, low quality

, etc.

```
--style
```
(optional): Style prompt keyword (the script will automatically add it to the front of the prompt), default is
```
anime
```
, options:
- ```
anime
```
  : Adds "anime style," prefix
- ```
flat illustration
```
  : Adds "flat illustration style," prefix
- ```
3d cartoon
```
  : Adds "3D cartoon style," prefix
- ```
sketch
```
  : Adds "sketch style," prefix
- ```
auto
```
  : Does not add style prefix (describe style manually in prompt)
- Note: wan2.6-t2i does not support the style API parameter, the script will integrate the style keyword into the prompt.
```
--size
```
(optional): Output resolution, default is
```
1280*1280
```
. Supported by wan2.6-t2i:
- Total pixels between 1280×1280 ~ 1440×1440
- Aspect ratio between 1:4 ~ 4:1
- Examples:
```
1280*1280
```
  (square),
```
768*2700
```
  (vertical long image),
```
1696*960
```
  (16:9 horizontal image)
```
--n
```
(optional): Number of images generated at one time, 1–4, default is 1.

--output-dir

(optional): Output directory, default is

./tmp/qwen-wanx-comic-<timestamp>

脚本工作流程

Script Workflow

使用
```
wan2.6-t2i
```
模型创建异步图像任务(新协议,不再支持 style API 参数,改为在 prompt 里融入风格描述)。
周期性轮询任务状态,直到
```
SUCCEEDED
```
或超时(默认 5 分钟)。
解析返回的
```
choices[].message.content[]
```
结构,下载图片 URL 到本地目录。
打印
```
MEDIA: <绝对路径>
```
行,方便 OpenClaw 自动将图片附在回复中。

Create an asynchronous image task using the
```
wan2.6-t2i
```
model (new protocol, no longer supports style API parameter, instead integrates style description into the prompt).
Periodically poll the task status until it is
```
SUCCEEDED
```
or times out (default 5 minutes).
Parse the returned
```
choices[].message.content[]
```
structure, download the image URL to the local directory.
Print the line
```
MEDIA: <absolute path>
```
to facilitate OpenClaw automatically attaching the image to the reply.

助手使用提示

Assistant Usage Tips

与用户对话时，优先将他们的需求转为 具体、画面感强的漫画描述，再调用本技能。
如果用户未明确说明风格，但提到“漫画、动漫、二次元、条漫、分镜”等关键词，可以主动选择本技能。
对于需要复杂长篇漫画、多页故事板的需求，只生成 代表性场景或封面，并告知用户当前技能以单图或少量场景为主。
生成完成之后告知用户。

When conversing with users, first convert their needs into specific, vivid comic descriptions before calling this skill.
If users do not explicitly specify the style but mention keywords such as "comic, anime, ACG, webcomic, storyboard", proactively select this skill.
For requests requiring complex long comics or multi-page storyboards, only generate representative scenes or covers, and inform users that this skill mainly focuses on single images or a small number of scenes.
Notify users after generation is completed.