← 返回
内容创作 Key 中文

seedream(doubao)-image-generation

Image generation via Volcengine Seedream API. Use this when you need to perform Text-to-Image (T2I), Image-to-Image (I2I), or general visual creative tasks.
通过火山引擎 Seedream API 进行图像生成。适用于需要进行文本到图像(T2I)、图像到图像(I2I)或一般视觉创意任务。
cp7553479
内容创作 clawhub v1.0.6 3 版本 100000 Key: 需要
★ 1
Stars
📥 1,024
下载
💾 153
安装
3
版本
#latest

概述

Seedream Image Generation Skill

This skill provides methods for calling the Volcengine Ark large model service platform's image generation models (Seedream).

Select Environment

The script provides implementations for both Python and JavaScript:

  • If your runtime is primarily Node.js/TypeScript or a web module, use scripts/seedream.js.
  • If your runtime involves complex data analysis, batch scheduling, or is purely Python, use scripts/seedream.py.

Supported Model Parameters (Model ID/Endpoint ID)

When invoking, the model parameter needs to be the deployment Endpoint ID from the Ark console (e.g., ep-202X...), which must be backed by one of the following foundational image generation models:

  • doubao-seedream-5-0-260128
  • doubao-seedream-4-5-251128
  • doubao-seedream-4-0-250828

Script Parameters

You must pass specific parameters to control the image generation. The available parameters are:

  • model (string, optional): Endpoint ID corresponding to the model. Defaults to doubao-seedream-5-0-260128.
  • prompt (string, required): Text description used to generate the image.
  • size (string, optional): Target image size. Supports "2K", "3K", "4K".
  • doubao-seedream-5-0 supports 2K and 3K.
  • doubao-seedream-4 and doubao-seedream-4-5 support 2K and 4K.
  • If not provided, this parameter is completely ignored in the API request.
  • You can also specify the desired resolution directly in the prompt. It is highly recommended to explicitly state the resolution in the prompt for the best results.
  • image (string | list of strings, optional): Local image path or image path list used for image-to-image generation. The scripts will read the local file, convert it to base64, and send it as the request image field.
  • watermark (boolean, optional): Whether to add a Volcengine AI watermark. Unprovided value assumes false.
  • optimize_prompt_options (object, optional): Configure automatic prompt optimization options. Format: JSON object. E.g., {"mode": "standard"}.
  • tools (list of objects, optional): Capabilities options for the model, formatted as JSON. Example: [{"type": "web_search"}].
  • output_format (string, optional): The output image format. Options include png, jpeg.
  • sequential_image_generation (string, optional): Sequential image generation strategy (e.g., "auto").
  • download_dir (string, optional): Local directory to save the generated images. If not provided, images will not be downloaded locally, and only the API response will be returned.

Recommended width and height values

ResolutionAspect RatioDimensions
:---------:-----------:---------
2K1:12048x2048
4:32304x1728
3:41728x2304
16:92848x1600
9:161600x2848
3:22496x1664
2:31664x2496
21:93136x1344
3K1:13072x3072
4:33456x2592
3:42592x3456
16:94096x2304
9:162304x4096
2:32496x3744
3:23744x2496
21:94704x2016
4K1:14096x4096
3:43520x4704
4:34704x3520
16:95504x3040
9:163040x5504
2:33328x4992
3:24992x3328
21:96240x2656

Usage Examples

Python Example:

python3 <current_skill_dir>/scripts/seedream.py \
  --prompt "A cute orange cat laying under the sun" \
  --model "ep-xxxxx..." \
  --size "1024x1024" \
  --watermark "false" \
  --output_format "png"

Node.js Example:

node <current_skill_dir>/scripts/seedream.js \
  --prompt "A cute orange cat laying under the sun" \
  --model "ep-xxxxx..." \
  --size "1024x1024" \
  --watermark "false" \
  --output_format "png"

python3 /scripts/seedream.py \

--prompt "Turn this product photo into a clean white-background ecommerce image" \

--image "/path/to/source.png" \

--model "ep-xxxxx..."

Example with Download Directory:

python3 <current_skill_dir>/scripts/seedream.py \
  --prompt "A beautiful sunset over the mountains" \
  --download_dir "/Users/vincent/Downloads/generated_images" \
  --model "ep-xxxxx..."

python3 /scripts/seedream.py \

--prompt "A futuristic cityscape" \

--model "ep-xxxxx..." \

--optimize_prompt_options '{"mode": "standard"}' \

--tools '[{"type": "web_search"}]'

Troubleshooting

If the scripts encounter issues, please check the following:

  • SEEDREAM_API_KEY: (env, required) Ensure this API key is set in your environment. It is the core authentication for Volcengine Ark.
  • SEEDREAM_BASE_URL: (env, optional) Ensure this is correct if you are using a custom endpoint. Defaults to https://ark.cn-beijing.volces.com/api/v3/images/generations.

For more detailed API documentation, please visit: https://www.volcengine.com/docs/82379/1541523?lang=zh

For more model id list or check the newest model id if required, please visit: https://www.volcengine.com/docs/82379/1330310?lang=zh#36969059, and then update this seedream-image-generation/SKILL.md file "Supported Model Parameters".

版本历史

共 3 个版本

  • v1.0.6 当前
    2026-05-01 01:26 安全 安全
  • v1.0.4
    2026-03-30 12:17
  • v1.0.5
    2026-03-18 19:18

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

content-creation

Humanizer

biostartechnology
消除AI写作痕迹,使文本更自然真实。基于维基百科"AI写作特征"指南,识别并修正夸张象征、宣传用语、肤浅-ing分析、模糊归因、破折号滥用、三项排比、AI词汇、负面平行结构及冗长连接词等模式。
★ 860 📥 199,942
content-creation

YouTube

byungkyu
使用托管OAuth集成YouTube Data API,支持搜索视频、管理播放列表、获取频道数据及评论互动,适用于用户需要时使用此技能。
★ 142 📥 41,084
content-creation

AdMapix

fly0pants
广告情报与应用数据分析助手,支持搜索广告素材、分析应用排名、下载量、收入及市场洞察,用于广告素材和竞品分析。
★ 295 📥 136,509