Chain
- Atrium
- Arbitrum + RH-Chain (when SDK)
- Cascade
- Solana
- August
- Solana
Benchmarks
Side-by-side against the closest comparables. Each row is a qualitative check against each protocol’s public docs and our own on-chain reads, not a head-to-head benchmark number. Where Atrium loses, we say so.
Margin compute
- Atrium
- Stylus (Rust, ~10-100x cheaper compute-heavy ops, per Arbitrum Stylus)
- Cascade
- Native Solana programs
- August
- N/A
Open adapter standard
- Atrium
- IPorticoAdapter v1.0 (MIT)
- Cascade
- Closed
- August
- Closed
Agent integration
- Atrium
- Sigil EIP-712 + Postern session keys + ERC-8004
- Cascade
- Limited
- August
- None
Formal verification
- Atrium
- 9 Kani proofs authored; CI lane Month 3
- Cascade
- Audit only
- August
- Audit only
UX polish today
- Atrium
- Foundation: Verifier Mode complete, app surfaces in progress
- Cascade
- More mature
- August
- More mature
Mainnet live
- Atrium
- No, testnet build phase
- Cascade
- Yes
- August
- Yes
| 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 | 9 Kani proofs authored; CI lane Month 3 | 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.