OpenVideo

Ollama for video models.
_{Run MiniMax H3 on your own GPU — install · pull · run — plus a drop-in skill so any coding agent can generate high-quality video.}

_{Ollama → local LLMs.  OpenVideo → local video.
v0.0.1 is exactly that loop for MiniMax H3 — not a multi-model platform yet. ▶ Watch the demo}

60-second start

# Linux / macOS — installs ComfyUI engine + pulls H3 weights (resumable, ~54 GB)
curl -fsSL https://open-video.ai/install | bash
Windows (PowerShell) — prefers WSL2 for the full H3 GPU path
irm https://open-video.ai/install.ps1 | iex
Same mental model as Ollama: pull → status → run
open-video pull h3                  # verify / resume H3 weights
open-video status                   # engine health + weight inventory (alias: ps)
open-video run "A lone astronaut planting a flag on a red dune at dusk" --duration 8
open-video "sunset waves" --dry-run # plan + validate, no GPU spent

| OS | Install | Generate | |---|---|---| | Linux | curl …/install \| bash | NVIDIA GPU · full H3 | | macOS | same curl (setup + dry-run) | H3 generation via community/MLX paths; not default | | Windows | irm …/install.ps1 \| iex | WSL2 for H3 GPU; native dry-run OK |

Hardware. Local-first; bring your own NVIDIA GPU. open-video recommend-quant picks the right weight tier for your card:

| VRAM | Quant tier | |---|---| | ≥ 22 GB | INT8 ConvRot (default, verified) | | 12–22 GB | INT8 + --lowvram offload | | 9–12 GB | W4 ConvRot (~10 GB) | | < 9 GB | NF4 (~8 GB entry) |

git clone https://github.com/open-video-ai/open-video && cd open-video
pip install -e .
open-video pull h3
open-video run "waves at sunset, golden hour" --duration 10 --model h3 --output out.mp4
ComfyUI at http://127.0.0.1:8188 (env OPEN_VIDEO_COMFYUI)

The agent path (what makes this different)

Point any agent host at the skill — it installs/pulls if needed, crafts the **official H3 3-field prompt**, validates against hard constraints, generates, and reviews:

| Skill | Use when | |---|---| | skill/h3-video/SKILL.md | v0.0.1 default — high-quality single/short H3 clips (T2V / I2V / FL2VA) | | skill/open-video/SKILL.md | Longer director path (plan → judge → stitch) — evolving |

Works with Claude Code, Cursor, Codex, OpenCode, and any host that loads SKILL.md. Quality is encoded, not left to chance: prompt grammar (backends/h3/PROMPT_GRAMMAR.md), a hard validator, and curated presets (open-video list-presets).

Three ways to use it

| Interface | For | Experience | |---|---|---| | 🤖 Skill harness | Any agent | Load skill/h3-video → agent generates H3 video end-to-end | | ⌨️ CLI | Developers / scripts | open-video pull · status · run (Ollama-shaped) | | 🖥️ Site | Discovery | open-video.ai — install + docs |

What works today vs what is designed next

| | v0.0.1 (shipped) | Designed (not wired yet) | |---|---|---| | Generate | Local MiniMax H3 via ComfyUI — pull / status / run | Multi-model backends (Wan, LTX, …) | | Agent path | skill/h3-video crafts official prompts + drives the CLI | Full multi-shot director agent | | Judge loop | Real VLM judge via env OPEN_VIDEO_VLM_URL/MODEL/KEY + automatic REFINE retries (OPEN_VIDEO_JUDGE_RETRIES, best take kept); honest PASS stub when unset | Best-of-N tournament judging | | Long film | Single clips (H3 shot length) | Planner → stitch multi-minute film | | Hosted try | Site /try is a browser mockup | Real hosted generate |

The generate → judge → refine loop runs today: point OPEN_VIDEO_VLM_URL at any OpenAI-compatible vision model and low-scoring shots regenerate automatically with a bumped seed (OPEN_VIDEO_JUDGE_RETRIES extra takes, best score kept — full take history in --json).

Why local

Closed tools charge per second and keep your prompts and footage in their pipeline. Open video models are now good enough to matter — what was missing is the simple local loop: install → pull → run, with best-practice prompting built in. v0.0.1 is that loop.

| | OpenVideo (local) | Typical closed SaaS | |---|---|---| | Model | MiniMax H3, open weights on your GPU | Vendor-hosted only | | Cost | Your GPU + electricity | Per-second API / subscription | | Data | Stays on your machine | Vendor pipeline | | Software license | Apache-2.0 | Proprietary ToS |

Licenses — read this before commercial use

• Detailed weights terms: docs/WEIGHTS_LICENSE.md.
• Code (this repo): Apache-2.0. Use it freely.
• Model weights are NOT covered by this repo's license. MiniMax H3 weights are distributed

• License-cleaner second backends (e.g. Wan) are on the roadmap.

How it compares (honest)

| | What | Open software? | Local open model? | Notes | |---|---|:--:|:--:|---| | OpenVideo | CLI + skill + H3 on ComfyUI | ✅ Apache-2.0 | ✅ H3 | this project — director/judge loop is scaffolding | | Runway | Closed SaaS | ❌ | ❌ | Hosted product | | Seedance | Closed agentic long video | ❌ | ❌ | Hosted product | | ComfyUI | Node-graph engine | ✅ GPL | via custom nodes | The runtime we drive — a dependency, not a competitor |

OpenVideo is not a foundation model and not a replacement for ComfyUI. v0.0.1 is the

install → pull → run layer plus an agent skill on top of H3.

Contributing

OpenVideo is a plugin surface — contribute what you're good at:

| You have | Contribute → | Effort | |---|---|---| | A great prompt | library/prompts/ — a verified recipe | 5 min | | A new model (Wan 2.2, Hunyuan, LTX) | backends// — a backend plugin | an afternoon | | A scoring method / vision judge | judges/ — a judge plugin | an afternoon | | A new engine (diffusers, SGLang) | engines// — an adapter | an afternoon | | A style LoRA | library/ — share it | 10 min |

See CONTRIBUTING.md for templates and GOVERNANCE.md for how decisions get made. We integrate, we don't reinvent — if a working project already does it, we wrap it as a plugin. Chat lands later; for now use GitHub Issues.

Architecture

Shipped path:

prompt / skill ──→ open-video CLI ──→ backends/h3 ──→ engines/comfyui ──→ mp4

Design target (modules exist as scaffolds; not all wired end-to-end):

concept ──→ planner → crafter → validator → backend → judge → stitcher → film

• backends/h3/ — MiniMax H3 plugin: prompt grammar, workflows, constraints.
• engines/comfyui/ — ComfyUI HTTP adapter (submit / wait / fetch).
• skill/h3-video/ — the agent harness.
• core/ — shared contracts + judge/planner scaffolding for later phases.

Status & roadmap

v0.0.1 — shipped: local H3 pull/run, agent skill harness, one-line installer, product site.

• Next: wire a real vision judge, multi-shot continuity, a license-clean second backend.
• Later, only when real: hosted generate, desktop packaging, community gallery.

Acknowledgments

Standing on the shoulders of open giants: ComfyUI (the engine), MiniMax H3 (the model), the woodfantasy prompt methodology (MIT-0), and VideoScore (judge direction). We integrate, not reinvent.

Security

Private vulnerability reporting: SECURITY.md.

License

Apache-2.0 © OpenVideo contributors · open-video.ai

_{OpenVideo · open-video.ai · Apache-2.0}