Ship custom AI models as easily as committing code.
Prompt Baking encodes prompt behavior directly into model weights. Fine-tune AI models with the convenience of prompt engineering.You shouldn’t need a PhD to teach your AI what it needs to know or wait for someone else’s model update to give you what you need. With Bread, your model permanently inherits the prompted behavior with zero input tokens at inference time.
The value: After baking, your model can exhibit prompted behavior with zero input tokens. The skills live in the weights, not in runtime prompts.
bgit is a git-native interface that makes AI model baking feel like regular software development. Version control your model weights with familiar git commands.
One file = One model (YAML configuration)
Commits = Model versions
Branches = Experiments
Git history = Complete audit trail
Understanding Baking
Learn the four-phase process to baking
Quickstart
Run a full bake with Python SDK
Guides
Step-by-step baking guides
Configuration
Python SDK configuration reference
Use the Python SDK for programmatic control over the baking process. Perfect for automation, integration, and advanced use cases.
Full programmatic control
Python-native API
Async support
Complete feature access
API Reference
Complete REST API documentation
API Playground
Try API endpoints interactively
Authentication
API key setup and security
Use the REST API for direct HTTP access to Bread’s baking platform. Perfect for custom integrations and non-Python environments.
