minotaur
SN112A decentralised DEX aggregator that finds the best swap routes across multiple exchanges
Every DeFi swap today goes through aggregators and market makers that extract fees while users accept whatever price they receive. Minotaur flips this by running a competitive solver network on Bittensor, where miners compete to find the best execution path for your swap intent and win fees only when they actually deliver. The subnet is up 40.0% over 30 days with 389 TAO in net inflows last week.
// Competitive swap solvers, on-chain
Minotaur (SN112) is a Bittensor subnet focused on swap-intent processing and execution optimization. Users submit signed swap intents, and a network of solver miners compete to find the best execution path using AMMs, RFQ market makers, and direct matches. Validators run Docker-based simulation epochs to score each solver on user surplus, gas efficiency, and correctness, then submit weights to the chain.
The simple version: Instead of accepting the first price an aggregator offers, a market of solvers compete to give you the best deal, and the winner earns fees only when they deliver.
Centralized equivalent: 1inch, Paraswap, CoW Protocol
How it works:
- Miners (solvers) receive swap intents, compute candidate settlements that maximize user surplus, and compete across AMMs, RFQ market makers, and direct matches. They earn solver fees on target chains and alpha token emissions based on performance, in a winner-takes-most model. Each submission must be signed by the solver's hotkey.
- Validators fetch pending orders from the aggregator, validate execution through Docker-based simulation epochs (default 5 minutes), compute miner scores from validation success rates, and submit weights to the Bittensor chain. Every validator processes the same event window and produces the same weight vector, ensuring deterministic scoring.
- The problem it solves: DeFi swaps are dominated by centralized aggregators and opaque market makers. Users have limited ability to verify they received the best available price, and solvers face no open competition incentive.
- The opportunity: Intent-based trading is an active area of development in DeFi. A subnet-native solver network that earns alpha token emissions while also earning real solver fees on target chains creates dual incentives for miners and better prices for users.
- The Bittensor advantage: Bittensor's incentive layer aligns solver competition with user outcomes. Deterministic scoring and cryptographic accountability via hotkey signing make manipulation harder than in off-chain solver networks.
- Traction signals: 389 TAO net inflow over 7 days. Price up 40.0% over 30 days. A CTO video explaining the intent-based design has circulated in the community, and SN112 has been highlighted as a standout subnet for agentic DeFi.
Category: Other - DeFi and Swap Optimization | Centralized Competitor: 1inch, Paraswap, CoW Protocol
Intent-based trading is gaining traction across DeFi as users increasingly want outcome guarantees (receive at least X tokens) rather than specifying exact execution paths. Minotaur applies this model on Bittensor, creating a subnet where solver competition is enforced by the network's incentive mechanism rather than by off-chain reputation systems.
Mechanism:
Users submit signed swap intents to the aggregator. Solver miners compute candidate settlements drawing on AMMs, RFQ market makers, aggregators, and direct matches, optimizing for user surplus as the primary scoring criterion, gas efficiency and correctness as secondary factors. The aggregator selects the best quote. Validators then run Docker-based simulation epochs to independently verify order execution results. Scoring is deterministic: every validator processes the same event window and produces the same weight vector. Weights are submitted to the Bittensor chain, and miners earn both on-chain solver fees on the target chain and alpha token emissions.
The roadmap targets better prices through solver competition, reduced user fees, and expanded cross-chain liquidity reach. GitHub has 38 commits from 3 contributors, with the last commit on 2026-01-27. The Gini coefficient of 0.783 indicates high stake concentration, worth monitoring as the subnet scales.
- Early-stage codebase: 38 GitHub commits with a last commit in January 2026 indicates relatively early development. Solver infrastructure maturity is uncertain.
- High stake concentration: A Gini of 0.783 means stake is concentrated among a small number of holders, which can create price volatility and governance imbalance.
- Execution complexity: Integrating AMMs, RFQ market makers, and direct matches across chains while maintaining deterministic validator scoring is technically demanding. Bugs in simulation epochs could produce incorrect weight vectors.
- Competitive landscape: Intent-based solving is an active research area with well-funded centralized competitors. Adoption depends on solvers finding participation worthwhile and users routing intents through the network.
Into the next one.
