Index of every page on this site: llms.txt
All skills
inference-sh avatar

/ai-video-generation

@c895a47
by inference.shinference-sh/skills819 stars
108
Run it yourself

Generate AI videos with Google Veo, Seedance 2.5, Kling, HappyHorse, Wan, Grok and 40+ models via inference.sh CLI. Models: Veo 3.1, Seedance 2.5, Kling v4, Runway Gen-4.5, FLUX 3 Video, HappyHorse 1.0, Wan 2.7, Grok Imagine Video 1.5, P-Video 2, OmniHuman, Fabric. Capabilities: text-to-video, image-to-video, reference-to-video, video editing, lipsync, avatar animation, video upscaling, foley sound. Use for: social media videos, marketing content, explainer videos, product demos, AI avatars. Triggers: video generation, ai video, text to video, image to video, veo, animate image, video from image, ai animation, video generator, generate video, t2v, i2v, ai video maker, create video with ai, runway, pika alternative, sora alternative, kling, seedance, happyhorse

Use this Skill: https://skilld.dev/gh/inference-sh/skills/ai-video-generation

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈198 tokens for metadata: the name and description. ≈1.7k when used: this file.

Description uses 9.9% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈792 characters, or 9.9%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Install the belt CLI skill: npx skills add belt-sh/cli

AI Video Generation

Generate videos with 40+ AI models via inference.sh CLI.

AI Video Generation

Quick Start

Requires inference.sh CLI (belt). Install instructions

belt login

# Generate a video with Veo
belt app run google/veo-3-1-fast --input '{"prompt": "drone shot flying over a forest"}'

Available Models

Text-to-Video

Model App ID Best For
Veo 3.1 Fast google/veo-3-1-fast Fast, with optional audio
Veo 3.1 google/veo-3-1 Best quality, frame interpolation
P-Video 2 pruna/p-video-2 Fast, economical, with audio support, up to 20s
WAN-T2V pruna/wan-t2v Economical 480p/720p
Wan 2.7 T2V alibaba/wan-2-7-t2v 720P/1080P, 2-15s
Grok Imagine Video 1.5 xai/grok-imagine-video-1-5 xAI, 1-15s, with audio
Seedance 2.5 bytedance/seedance-2-5 Text/image/ref-to-video with sync audio, up to 1080p, 4-30s
Kling v4 klingai/video-v4 Cinematic, up to 1080p, up to 30s
Runway Gen-4.5 runway/gen-4-5 Runway's latest, 2-10s
FLUX 3 Video bfl/flux-3-video BFL, keyframes, with audio
HappyHorse T2V alibaba/happyhorse-1-0-t2v Physically realistic, up to 15s

Image-to-Video

Model App ID Best For
Wan 2.7 I2V alibaba/wan-2-7-i2v High quality i2v, first+last frame
WAN-I2V pruna/wan-i2v Economical 480p/720p
P-Video 2 pruna/p-video-2 Fast i2v with audio
Seedance 2.5 bytedance/seedance-2-5 Animate images with sync audio, up to 1080p
HappyHorse I2V alibaba/happyhorse-1-0-i2v Animate images, up to 1080P/15s
HappyHorse R2V alibaba/happyhorse-1-0-r2v Character-preserving from references

Avatar / Lipsync

Model App ID Best For
OmniHuman 1.5 bytedance/omnihuman-1-5 Multi-character
OmniHuman 1.0 bytedance/omnihuman-1-0 Single character
Fabric 1.0 falai/fabric-1-0 Image talks with lipsync
PixVerse Lipsync falai/pixverse-lipsync Realistic lipsync

Video Editing

Model App ID Best For
HappyHorse Edit alibaba/happyhorse-1-0-video-edit Natural language video editing

Utilities

Tool App ID Description
MMAudio infsh/mmaudio Add sound effects to video
Topaz Upscaler falai/topaz-video-upscaler Upscale video quality
Media Merger infsh/media-merger Merge videos with transitions

Browse All Video Apps

belt app list --category video

Examples

Text-to-Video with Veo

belt app run google/veo-3-1-fast --input '{
  "prompt": "A timelapse of a flower blooming in a garden"
}'

Grok Video

belt app run xai/grok-imagine-video-1-5 --input '{
  "prompt": "Waves crashing on a beach at sunset",
  "duration": 5
}'

Image-to-Video with Wan 2.7

belt app run alibaba/wan-2-7-i2v --input '{
  "first_frame": "https://your-image.jpg",
  "prompt": "slow camera push-in"
}'

AI Avatar / Talking Head

belt app run bytedance/omnihuman-1-5 --input '{
  "image": "https://portrait.jpg",
  "audio": "https://speech.mp3"
}'

Fabric Lipsync

belt app run falai/fabric-1-0 --input '{
  "image": "https://face.jpg",
  "audio": "https://audio.mp3"
}'

Seedance 2.5 Text-to-Video with Audio

belt app run bytedance/seedance-2-5 --input '{
  "prompt": "a jazz band performing in a dimly lit club",
  "generate_audio": true,
  "duration": 10
}'

Seedance 2.5 Image-to-Video

belt app run bytedance/seedance-2-5 --input '{
  "image": "https://your-image.jpg",
  "prompt": "gentle camera movement, leaves rustling in the wind",
  "generate_audio": true
}'

Seedance 2.5 Reference-to-Video

belt app run bytedance/seedance-2-5 --input '{
  "prompt": "The person in @Image1 walking through a garden",
  "reference_images": ["https://portrait.jpg"],
  "generate_audio": true
}'

HappyHorse Text-to-Video

belt app run alibaba/happyhorse-1-0-t2v --input '{
  "prompt": "a golden retriever running through autumn leaves, slow motion",
  "duration": 10,
  "resolution": "1080P"
}'

HappyHorse Video Editing

belt app run alibaba/happyhorse-1-0-video-edit --input '{
  "video": "https://your-video.mp4",
  "prompt": "change the background to a snowy mountain landscape"
}'

PixVerse Lipsync

belt app run falai/pixverse-lipsync --input '{
  "video": "https://talking-head.mp4",
  "audio": "https://speech.mp3"
}'

Takes a video, not a still image. Omit audio and pass text (plus optional voice_id) to use the built-in TTS.

Video Upscaling

belt app run falai/topaz-video-upscaler --input '{"video": "https://..."}'

Add Sound Effects (Foley)

belt app run infsh/mmaudio --input '{
  "video_input": "https://silent-video.mp4",
  "prompt": "footsteps on gravel, birds chirping"
}'

Merge Videos

belt app run infsh/media-merger --input '{
  "media_files": [
    {"file": "https://clip1.mp4", "transition_type": "crossfade"},
    {"file": "https://clip2.mp4"}
  ]
}'

Related Skills

# Full platform skill (all apps)
npx skills add inference-sh/skills@infsh-cli

# Pruna P-Video 2 (fast & economical)
npx skills add inference-sh/skills@p-video

# Google Veo specific
npx skills add inference-sh/skills@google-veo

# Seedance 2.5
npx skills add inference-sh/skills@seedance

# HappyHorse 1.0
npx skills add inference-sh/skills@happyhorse

# AI avatars & lipsync
npx skills add inference-sh/skills@ai-avatar-video

# Text-to-speech (for video narration)
npx skills add inference-sh/skills@text-to-speech

# Image generation (for image-to-video)
npx skills add inference-sh/skills@ai-image-generation

# Twitter (post videos)
npx skills add inference-sh/skills@twitter-automation

Browse all apps: belt app list

Documentation

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts2d5 checks · Risk SAFE
  • Gen Agent Trust Hub3d

    This skill provides instructions and examples for using the inference.sh CLI tool ('belt') to generate AI videos across various models. It uses official vendor resources and follows standard practices for a documentation-focused skill.

  • Socket2d

    No alerts

  • Snyk3d

    Risk: LOW · No issues

  • Runlayer8mo

    1/1 file flagged

  • ZeroLeaks7mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at c895a47. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 3 days ago.

Activeupdated 4 days ago
What it can do
Runs commands
All 1 allowed tools
Bash(belt *)
  • video-generation
  • text-to-video
  • image-to-video
  • lipsync
  • avatar-animation
  • veo
  • seedance
  • happyhorse
  • inference-sh

README badge

README badge for inference-sh/skills/ai-video-generation

Generate videos with 40+ AI models (Veo, Seedance, HappyHorse, Wan, Grok) via the inference.sh CLI, supporting text-to-video, image-to-video, avatar animation, lipsync, and video editing. Use for social media content, product demos, and marketing videos where you need fast or high-quality output across multiple model providers.

Generated from the current SKILL.md.

What video models does this skill support?
40+ models including Google Veo 3.1, Seedance 2.0, HappyHorse, Wan 2.5, Grok Video, and others. Each model has specific strengths: Veo for quality, Seedance for audio sync, HappyHorse for realistic physics, Wan for image animation.
Does this skill generate audio for videos?
Some models like Seedance 2.0 and Veo 3 support optional audio generation. You can also add sound effects via HunyuanVideo Foley or use lipsync models (Fabric, OmniHuman, PixVerse) to sync existing audio to avatars.
Can I animate still images into videos?
Yes. Use image-to-video models like Wan 2.5, Seedance 2.0, or HappyHorse I2V. Seedance 2.0 R2V can preserve character details from reference images.
Do I need to install the belt CLI separately?
Yes. The skill requires the inference.sh CLI (`belt`). Install it first with `npx skills add belt-sh/cli`, then authenticate with `belt login`.
What resolution and duration limits do these models have?
Most models support up to 1080p. HappyHorse supports up to 15 seconds, Veo 3.1 and Grok Video are configurable, and Seedance 2.0 supports up to 10-second durations by default.

Generated from the current SKILL.md. These answers refresh after source changes.