Skip to main content

animal-mediakit

AI 图像/视频生成 + 本地图像处理 CLI 工具,通过 animal-gateway 统一调用多家 AI 供应商模型。

快速开始

cd extensions/animal-mediakit
uv sync
uv run animal-mediakit auth login
uv run animal-mediakit auth ping

供应商支持矩阵

能力 Gemini OpenAI Volcengine DashScope 腾讯云 VOD
生图 gemini-2.5-flash-image gpt-image-1.5 / azure/gpt-image-2 / tencent/gpt-image-2 Seedream 5.0 Wan 2.6 GEM/SI/Kling/Vidu/Jimeng/Hunyuan/Qwen/OG(gpt-image-2)
生视频 Veo 2.0 Sora 2 Seedance 2.0 Wan 2.6 T2V Kling/Vidu/Hailuo/Seedance/GV/OS
图生视频 - - - - Kling/Vidu/GV (首帧/参考)
图片编辑 - gpt-image-1.5 - - -
图片分析 Gemini Vision GPT-5 Vision - Qwen3-VL -
对口型 - - - - Kling (lip_sync)
数字人 - - - - Kling (avatar_i2v)
场景化生图 - - - - 电商场景
超分增强 - - - - ProcessMedia

CLI 命令参考

认证

animal-mediakit auth login              # SSO 登录
animal-mediakit auth logout             # 登出
animal-mediakit auth status             # 查看当前认证状态
animal-mediakit auth ping               # 测试 gateway 连通性

AI 图像生成

animal-mediakit generate image "prompt" -o out.png --model gemini/gemini-2.5-flash-image
animal-mediakit generate image "prompt" -o out.png --model openai/gpt-image-1.5
animal-mediakit generate image "prompt" -o out.png --model azure/gpt-image-2
animal-mediakit generate image "prompt" -o out.png --model tencent/gpt-image-2
animal-mediakit generate image "prompt" -o out.png --model doubao-seedream-5-0-260128
animal-mediakit generate image "prompt" -o out.png --model dashscope/wan2.6-image
animal-mediakit generate image "prompt" -o out.png --model vod/gem-3.1
animal-mediakit generate image "prompt" -o out.png --model vod/kling-3.0
animal-mediakit generate image "prompt" -o out.png --model vod/og-image2_low
animal-mediakit generate image "prompt" -o out.png --model vod/og-image2_high

AI 视频生成

animal-mediakit generate video "prompt" -o out.mp4 --model gemini/veo-2.0-generate-001
animal-mediakit generate video "prompt" -o out.mp4 --model openai/sora-2.0
animal-mediakit generate video "prompt" -o out.mp4 --model doubao-seedance-2-0-260128
animal-mediakit generate video "prompt" -o out.mp4 --model vod/kling-3.0 --duration 5
animal-mediakit generate video "prompt" -o out.mp4 --model vod/gv-3.1 --audio

VOD 高级场景

# 图生视频(首帧参考)
animal-mediakit generate video "prompt" -o out.mp4 --model vod/kling-3.0 --from-image ref.jpg

# 对口型
animal-mediakit generate lipsync --video input.mp4 --audio voice.mp3 -o out.mp4

# 数字人
animal-mediakit generate avatar --image person.png --audio speech.mp3 -o out.mp4

# 场景化生图(电商)
animal-mediakit generate scene-image "product on marble" --model vod/gem-3.0 -o out.png

# 超分增强
animal-mediakit enhance --file-id xxx -o enhanced.mp4

自定义主体

animal-mediakit element create --name "my-cat" --description "orange tabby" --image https://example.com/cat.jpg

图片编辑

animal-mediakit generate edit photo.jpg --prompt "把天空变成日落" -o edited.jpg
animal-mediakit generate edit photo.jpg --prompt "移除背景" -o edited.png --mask mask.png

图片分析

animal-mediakit analyze photo.jpg --prompt "描述这张图片"
animal-mediakit analyze photo.jpg --model openai/gpt-5 --prompt "这是什么品种的猫?"

本地图像处理

animal-mediakit image info photo.jpg
animal-mediakit image crop photo.jpg --rect 100,100,500,400 -o cropped.jpg
animal-mediakit image resize photo.jpg --width 800 -o resized.jpg
animal-mediakit image flip photo.jpg --direction horizontal -o flipped.jpg
animal-mediakit image rotate photo.jpg --angle 90 -o rotated.jpg
animal-mediakit image convert photo.jpg --format webp -o photo.webp
animal-mediakit image compress photo.jpg --quality 80 -o compressed.jpg

环境变量

变量 说明 默认值
MEDIAKIT_DEFAULT_IMAGE_MODEL 默认图像生成模型 gemini/gemini-2.5-flash-image
MEDIAKIT_DEFAULT_VIDEO_MODEL 默认视频生成模型 gemini/veo-2.0-generate-001
MEDIAKIT_HTTP_MAX_RETRIES HTTP 请求最大重试次数 3
MEDIAKIT_HTTP_RETRY_DELAY 重试基础延迟(秒) 2.0

测试

uv run pytest tests/ -x                          # 单元测试
uv run pytest tests/ -m integration -v -s        # 集成测试(需要有效凭证)

详细文档

Release files for animal-mediakit 1.18.65

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for animal-mediakit 1.18.65
File Size Uploaded
animal_mediakit-1.18.65.tar.gz 26.7 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for animal-mediakit 1.18.65
File Interpreter ABI Platform
animal_mediakit-1.18.65-py3-none-any.whl Python 3 none any Details

Total release size: 63.8 kB

Release files / animal_mediakit-1.18.65.tar.gz

Download URL animal_mediakit-1.18.65.tar.gz
Size 26.7 kB
Tags Source
SHA-256 checksum
How to use checksums
dbe2e511120e2bdd1a8810c5b5e0825367b78525f5ef412f04c4b6479bf89551
BLAKE2b-256 checksum
How to use checksums
a46d3f44a1f6382a80542334086975570138f062c72b5d75b21b62eafb76c4df
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.13.12

Release files / animal_mediakit-1.18.65-py3-none-any.whl

Download URL animal_mediakit-1.18.65-py3-none-any.whl
Size 37.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
4de12c13da03a6d78d5303cb723fea007d969dc994a78aa8ea99da4c8e399d40
BLAKE2b-256 checksum
How to use checksums
0b070bf04369b08c1aa46e645cab8dfd8916404f80599ba82ecdf76210752ed5
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.13.12

Release history Release notifications | RSS feed

This release

1.18.65 This release

2 release files

1.3.62

2 release files

1.2.39

2 release files

1.2.38

2 release files

1.2.37

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page