Benchmarks
Side-by-side against the closest comparables. Numbers come from competitor docs and on-chain reads. Where Atrium loses, we say so.
| Dimension | Atrium | Cascade (Solana) | August (Solana) |
|---|---|---|---|
| Chain | Arbitrum + RH-Chain (when SDK) | Solana | Solana |
| Margin compute | Stylus (Rust, ~10-100x cheaper compute-heavy ops, per Arbitrum Stylus) | Native Solana programs | N/A |
| Open adapter standard | IPorticoAdapter v1.0 (MIT) | Closed | Closed |
| Agent integration | Sigil EIP-712 + Postern session keys + ERC-8004 | Limited | None |
| Formal verification | 5 Kani+proptest invariants in CI | Audit only | Audit only |
| UX polish today | Foundation: Verifier Mode complete, app surfaces in progress | More mature | More mature |
| Mainnet live | No, testnet build phase | Yes | Yes |
Honest read: the Solana-native comparables have UX polish and mainnet liveness Atrium does not yet match. Atrium leads on chain choice (EVM is larger), open standards, agent-layer depth, and formal verification. The polish gap closes as the app surfaces complete; the standards gap holds.