Bittensor

About Bittensor

Bittensor organises machine-learning work into subnets where participants produce outputs and are scored by validators, with emissions flowing to those judged most useful. The intent is a market for model inference and training that does not depend on a single provider. Whether peer scoring reliably tracks real usefulness is an open question, and the design continues to change.

Key features

• Subnets specialised to particular ML tasks
• Validator scoring determines reward distribution
• Incentives for inference and training contribution
• No single provider controls model access

Supported Networks

• Multi-chain

Tags

• Staking
• Infrastructure
• AI