Each challenge defines a dataset, metric, and baseline that miners compete to beat. When a model surpasses the baseline on hidden tests, validators confirm the result and trigger a payout. It's how research becomes measurable progress: open, reproducible and rewarded only on proof.
Autoresearch is the current BitSota challenge path: live tasks publish repositories, metrics, baselines, and validator replay rules so miners can produce measurable AI progress.
What is Autoresearch?
Autoresearch turns open research tasks into measurable competitions. Miners submit improvements against a defined metric, validators replay the work, and rewards follow only confirmed progress.
The Traditional Approach
Manual Research Loops
Researchers choose a direction, run experiments, and manually decide what to trust
Private Evaluation
Results often depend on local scripts, private data, or one-off review
Unpaid Dead Ends
Compute can be spent without a confirmed result or reusable artifact
Autoresearch Approach
Task-Defined Research
Each task publishes the repository, metric, baseline, and allowed submission surface
Validator Replay
Independent validators rerun candidate work and score the observed result
Proof-Gated Rewards
Accepted improvements flow into reward snapshots, Pool claims, and auditable payout proofs
Start Mining
Set-up mining in less than 60 seconds. One-click CPU/GPU
System Requirements
