[ DOCUMENTATION ]
Otter documentation
Learn how to compose transformer architectures visually and export production-ready PyTorch code.
[ GUIDES ]
[ 01 ][ → ][ 02 ][ → ][ 03 ][ → ]
Getting Started
[ 5 min read ]Compose your first transformer architecture in Otter. From canvas to exported PyTorch in under five minutes.
Component Reference
[ Reference ]The full registry: embeddings, attention, FFN, normalization, routing. What each block accepts, what it emits.
Training Guide
[ 10 min read ]Run your exported model on local GPU, Google Colab, or Lambda Labs.
[ QUICK START ]
- 01. Open the Canvas →
- 02. Drag an Embedding component onto the canvas
- 03. Add Attention, FFN, and Output components
- 04. Connect them in topological order
- 05. Click Export when the architecture validates