A Claude Code skill that deploys a self-hosted LTX-2.3 (22B) video model to the user's own Modal serverless GPU account and drives it from chat: text-to-video, image-to-video, keyframe interpolation, video-to-video, and IC-LoRA control, with synced audio. Calls the deployed Modal app directly, billed per second of GPU time.
STACK
LTX-2.3ModalPython
“Own your AI video pipeline. LTX-2.3 (22B) self-hosted on your Modal GPU via a Claude Code skill — t2v, i2v, keyframes, v2v + synced audio. ~$0.02 per 5s clip, idle = $0.”
agent-skillsai-videoclaude-code-skillclaude-skillsdiffusion-modelsimage-to-videoltx-videomodalself-hostedserverless-gputext-to-videovideo-generation
About Frameworks & toolkits
Rendering frameworks and all-in-one kits: the layer that turns code an agent writes into frames, plus the skills that teach the agent how to drive it.