generative-video

The open .char format for consistent characters. Face, body, wardrobe & voice in one file, consistent across every model.

AI-native video editing built around real creative timelines, generative workflows, and local agent control.

Run MiniMax Hailuo H3 and LTX-2.5 video generation locally on a Mac. Joint audio+video, character LoRA training, one-click Pinokio install. MLX — no CUDA, no cloud, no API key.

Open-source Next.js SaaS for Seedance 2.0 , Seedance 2.5 and Seedance 2 Mini video generation — Stripe billing, credits, NextAuth, and Prisma out of the box.

Open-source local implementation of MiniMax H3 Context-IR. Turn simple prompts into structured, validated H3 video briefs.

A curated collection of AI video prompting resources, featuring official guides, prompt templates, cinematic techniques, and audio-visual synchronization methods for Veo, Sora, Runway, Pika, Kling, and other leading AI video tools.

Realtime interruptible science tutoring with continuously generated H3 Max video.

Open Film Skills: 21 independent modules for story, directing, visual assets and AI-film production, with research galleries and playable 3D previs.

Hermes Agent skill for typography-led lyric videos using GPT Image 2, Gemini Omni Flash, and exact-text QC

Four ComfyUI nodes for MiniMax H3: name your media with @, lock exact dialogue, and get the structured brief H3 wants. No <Picture 1> bookkeeping, no second service to start.

VS Code-inspired rich H3 prompt editor and reference palette for ComfyUI

Open-source Veo video generator — production-ready Next.js SaaS for text-to-video, image-to-video, and multi-image editing powered by Google Veo. Stripe billing, credits, NextAuth, and Prisma out of the box.

Beyond Time Shifts (ECCV 2026): a reference-free audio-visual synchronization evaluator built on Qwen2.5-Omni-3B, with SynthSync data, R-GRPO training, and the SyncBench benchmark.

Advanced, end-to-end, enterprise-grade agentic AI pipeline that automates competitor ad intelligence, performs multimodal creative strategy extraction, enables brand-safe adaptation, and generates AI video ads using LLM reasoning, multimodal analysis, and deterministic workflow orchestration with full auditability.

Agent skill + template: short videos made entirely from code. One HTML file, every frame a pure function of (frame, seed, width). Claude Code, Codex, Gemini CLI, Cursor.

A structured, JSON-based schema for precise, multi-scene AI video prompting and cinematic control.

334 verified prompts people used to make videos with Claude Opus 5.5, copied word for word from each creator's post. Previews, EN/中文 translations, and a director's toolkit.

Motion graphics as code. An agent skill for building frame-exact animated logos, kinetic typography and brand videos as a single HTML file, then exporting them to MP4 in 1:1, 4:5 and 9:16.

Turn a story into a finished film — on your own Mac, no cloud. Story Forge is a local-only generative cinema pipeline: stills, motion, voices, music, grade, and sound, start to finish. Flux · LTX-2 · Piper · ACE-Step · ffmpeg — built for and running 100% on a Mac (M5, 128GB).

Advanced, end-to-end, enterprise-grade agentic AI pipeline that automates competitor ad intelligence, performs multimodal creative strategy extraction, enables brand-safe adaptation, and generates AI video ads using LLM reasoning, multimodal analysis, and deterministic workflow orchestration with full auditability.