incocreativedev/tessera-core

An activation-based protocol for AI-to-AI knowledge transfer across architectures

36
/ 100
Emerging

This project helps AI researchers and machine learning engineers transfer learned knowledge between neural networks, even if they have entirely different architectures. It takes a trained model and an untrained model, along with training data, and produces a 'Tessera Token' — a compact, self-describing binary file that represents the transferred knowledge. This allows models to efficiently teach each other without needing to retrain from scratch or share their internal weights.

Used by 1 other package. Available on PyPI.

Use this if you need to quickly transfer the 'understanding' from one AI model to another, perhaps to jumpstart training on a new architecture or reuse knowledge across different model types.

Not ideal if you're looking for a simple model serialization tool or if your primary goal is to combine model weights directly.

AI-knowledge-transfer neural-network-training model-distillation deep-learning-research architecture-agnostic-transfer
Maintenance 10 / 25
Adoption 6 / 25
Maturity 20 / 25
Community 0 / 25

How are scores calculated?

Stars

11

Forks

Language

Python

License

Last pushed

Feb 25, 2026

Commits (30d)

0

Dependencies

5

Reverse dependents

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/incocreativedev/tessera-core"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.