Skip to content
SOMA

Network

SOMA is a blockchain, because it relies on trust.

Three requirements demand it:

  1. Issuance: The network needs to mint and distribute $SOMA according to rules no single party controls.
  2. Settlement: Bonds and rewards require a shared ledger that everyone trusts.
  3. Verification: Claims must be auditable by anyone, not just a central authority.

A database could handle the data. Only a blockchain can handle the trust.

Epochs

Section titled “Epochs”

An epoch is one day — the network’s heartbeat. Each epoch, the cycle repeats:

  • Validators generate targets
  • Submitters download weights and submit data
  • Challenge window runs on previous epoch’s winners
  • Rewards are distributed

Then it starts again. Epochs create predictable rhythm. Submitters know when to submit. Models know when weights lock. The entire network synchronizes around this cycle.

Consensus

Section titled “Consensus”

SOMA is built on Sui’s consensus layer. Everything else — Sui’s execution environment, virtual machine, smart contract system — is removed. What remains is Mysticeti, a consensus mechanism optimized for speed:

  • Transactions confirm in under a third of a second
  • Throughput exceeds 200,000 transactions per second

This matters because SOMA’s operations are lightweight but frequent. Submissions, verifications, challenges — thousands per epoch. The consensus layer must keep up without becoming a bottleneck.

Collusion Resistance

Section titled “Collusion Resistance”

No single anti-cheating mechanism is sufficient. SOMA’s design layers several:

  • Bonds make false claims expensive.
  • Optimistic verification keeps costs low while maintaining security.
  • Commit-reveal prevents models from copying each other mid-round.

But the deeper protection is structural: the system separates knowledge from action at every step.

  • Models must publish weights before targets are known — so a model developer who is also a submitter can’t tune their model to favor their own submissions.
  • Models are assigned to targets via stake-weighted KNN over their embeddings — assignment is deterministic based on model specialization and stake, not manipulable.
  • Targets are generated via VDF from consensus checkpoints — so no validator can predict or manipulate them alone.

Each mechanism addresses a different attack vector. Together, they make honest participation the profitable strategy.