Pricing
Pay for the providers, not the pixels
Rendering is Manim on your own hardware, so it costs nothing to run. The only real spend is inference — whether you point it at a local model or a hosted one.
Local
Freeself-hosted
Everything runs on your machine. Ollama writes, gTTS speaks, Whisper measures. No key, no bill, no upload.
Start free- Unlimited videos
- Up to 1080p60
- Ollama for script and scene code
- gTTS narration
- Local Whisper word timings
- Burned karaoke captions
- Voice selection
- Hosted rendering
Studio
$24per month
Bring hosted providers. ElevenLabs returns its own word timings, so transcription is skipped and jobs finish a stage sooner.
Start free trial- Everything in Local
- ElevenLabs voices with word timings
- Nine-stage pipeline (no transcription pass)
- OpenAI-compatible script generation
- Up to 4K renders
- Script library and reusable lexicons
Department
Talk to usannual
For faculties and teams rendering at volume. Distributed rendering across machines, and a shared style and lexicon baseline.
Get in touch- Everything in Studio
- Distributed scene rendering
- Shared theme and lexicon presets
- Priority render queue
- Bring your own storage bucket
What actually costs money
Script + scene generation
Two LLM passes per video: one to write the script, one per beat for the animation. About 20k tokens for a ninety-second explainer.
Narration
gTTS is free. ElevenLabs runs around $0.30 per thousand characters — roughly nine cents for a ninety-second script.
Rendering
Manim on your CPU or GPU. A ninety-second 1080p60 video takes two to three minutes and costs nothing but electricity.