Running a new Distributed Hash Table (DHT) is a way to create a new network to host a machine-learning model. This is meant for those planning on creating a voting process to initialize a new model or multiple new models to the blockchain.

Feel free to run a DHT node but DHT nodes do not garner rewards. They are strictly meant to connect to a set of peers.


python -m petals_tensor.cli.run_dht

Voting is not implemented in v1.0

Voting in a new model requires hosting the model with the required minimum amount of peers and having enough balance to cover the fee to add the model to the blockchain if voted in successfully.

If you are looking to become a model validator (peer) on an existing AI model, read here.

Documentation coming post testnet v1.0.

Last updated