ZenMux Media Gen

v1.0.0

Generate images & videos with ZenMux. Support multiple image models (Gemini, Qwen, Hunyuan, etc.) and video models (Veo, Seedance) via one API key.

0· 172· 1 versions· 0 current· 0 all-time· Updated 14h ago· MIT-0

Install

openclaw skills install zenmux-media-gen

ZenMux Media Gen 🎬

用 ZenMux API 一把钥匙生成图片视频

支持的图片模型:

  • google/gemini-3-pro-image-preview(Gemini 3 Pro Image, 高质量)
  • google/gemini-2.5-flash-image(Gemini 2.5 Flash Image,性价比高)
  • qwen/qwen-image-2.0(通义万相 Image 2.0)
  • tencent/hunyuan-image3(腾讯混元 Image 3)
  • openai/gpt-image-1.5(OpenAI GPT Image)

支持的视频模型:

  • google/veo-3.1-generate-001(Google Veo 3.1,高质量)
  • volcengine/doubao-seedance-1.5-pro(字节Seedance)

🔥 你可以做什么

图片生成

"生成一张赛博朋克风格的城市夜景,霓虹灯,雨夜,电影感"

视频生成(Veo 3.1)

"生成5秒视频:一个宇航员在火星上漫步,日落,电影感"

Quick Start

export ZENMUX_API_KEY="your-key"

🖼️ Image Generation

Endpoint

  • Base URL: https://api.zenmux.ai/v1
  • POST /chat/completions(OpenAI兼容格式)

curl 示例

curl -X POST "https://api.zenmux.ai/v1/chat/completions" \
  -H "Authorization: Bearer $ZENMUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-2.5-flash-image",
    "messages": [{"role": "user", "content": "A cute red panda, ultra-detailed, cinematic lighting"}],
    "modalities": ["image"]
  }'

可用图片模型

模型ID描述价格
google/gemini-3-pro-image-previewGemini 3 Pro Image,高质量$12-18/M
google/gemini-2.5-flash-imageGemini 2.5 Flash,性价比高$2.5/M
google/gemini-3.1-flash-image-previewGemini 3.1 Flash$3/M
qwen/qwen-image-2.0通义万相 Image 2.0-
qwen/qwen-image-2.0-pro通义万相 Image 2.0 Pro-
tencent/hunyuan-image3腾讯混元 Image 3-
openai/gpt-image-1.5OpenAI GPT Image 1.5$10/M
volcengine/doubao-seedream-5.0-lite字节Seedream 5.0$0.032/counts
z-ai/glm-image智谱GLM Image-
klingai/kling-v2可灵AI$0.014/counts

🎞️ Video Generation (Veo 3.1)

Create Task

  • Base URL: https://api.zenmux.ai/v1
  • POST /images/generations(Google Veo 使用 Image Generation API)
curl -X POST "https://api.zenmux.ai/v1/images/generations" \
  -H "Authorization: Bearer $ZENMUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/veo-3.1-generate-001",
    "prompt": "A astronaut walking on Mars, sunset, cinematic",
    "parameters": {
      "duration": 8,
      "resolution": "720p"
    }
  }'

Poll Task

curl "https://api.zenmux.ai/v1/images/generations/{task_id}" \
  -H "Authorization: Bearer $ZENMUX_API_KEY"

可用视频模型

模型ID描述价格
google/veo-3.1-generate-001Google Veo 3.1,8秒高质量视频$0.4-0.6/秒
volcengine/doubao-seedance-1.5-pro字节Seedance 1.5 Pro,音视频同步$2.33/M

Python Client

# 生成图片(默认使用 Gemini 3 Pro Image)
python3 {baseDir}/scripts/zenmux_media_client.py image \
  --prompt "A cute red panda, cinematic lighting" \
  --out "out.png"

# 生成视频任务(Veo 3.1)
python3 {baseDir}/scripts/zenmux_media_client.py video-create \
  --model "google/veo-3.1-generate-001" \
  --prompt "A astronaut walking on Mars, sunset" \
  --duration 8

# 轮询任务状态
python3 {baseDir}/scripts/zenmux_media_client.py video-status --task-id YOUR_TASK_ID

# 等待直到成功并自动下载
python3 {baseDir}/scripts/zenmux_media_client.py video-wait --task-id YOUR_TASK_ID --download --out out.mp4

环境变量

变量名必填说明
ZENMUX_API_KEYZenMux API Key,从 https://zenmux.ai 获取

配置示例

使用 Gemini 3 Pro(高质量,默认)

export ZENMUX_API_KEY="your-key"
python3 scripts/zenmux_media_client.py image \
  --prompt "高清电影质感海报" \
  --out movie_poster.png

使用 Veo 3.1(视频)

python3 scripts/zenmux_media_client.py video-create \
  --model "google/veo-3.1-generate-001" \
  --prompt "宇航员在火星漫步" \
  --duration 8

Version tags

latestvk977c31jac70q9zmk9rpph3vwn832fkd

Runtime requirements

🎬 Clawdis
Binspython3, curl
EnvZENMUX_API_KEY
Primary envZENMUX_API_KEY