engine-manim

AI-assisted Manim scene generation with a Next.js UI and server-side Python rendering.

Docker-First Quickstart (Recommended)

1) Configure env

cp .env.example .env.local

Set GROQ_API_KEY in .env.local.

2) Run in development

docker compose up --build

Open http://localhost:3000.

The app runs with hot reload and writes render artifacts under:

• python/jobs
• public/jobs

3) Stop

docker compose down

If you also want to remove dev volumes (including cached dependencies):

docker compose down -v

Production Container

Build:

docker build -t engine-manim:latest .

Run:

docker run --rm -p 3000:3000 --env-file .env.local engine-manim:latest

Optional Host-Native Setup (Without Docker)

You need Node.js + Python + Manim system dependencies (FFmpeg/LaTeX stack), then:

npm ci
npm run dev

Troubleshooting

Missing GROQ_API_KEY

• Symptom: generation endpoint fails quickly.
• Fix: ensure .env.local exists and includes a valid GROQ_API_KEY, then restart the container.

Render failures inside container

• Symptom: /api/generate or /api/compile returns render errors.
• Fix: inspect container logs with docker compose logs -f app and verify the generated Manim code is valid.

Slow or failing renders

• Symptom: long render times, process crashes, or unstable behavior.
• Fix: allocate more Docker memory/CPU, reduce scene complexity, and retry.

Security Note

• Never commit .env.local.
• Rotate any key that was accidentally shared in commits, screenshots, terminal history, or chat.