Abstract

Hybrid RFQ + AMM execution with practical privacy, built for high‑intent flow on Solana.

Pre‑launch
Privacy‑forward
Hybrid routing

Velaris is a non‑custodial DEX that pairs low‑latency RFQ price discovery against market makers with AMM aggregation for deterministic on‑chain depth. We minimize trivial timing and linkage across the trade lifecycle by isolating sessions, adding controlled jitter, probabilistically batching settlement, and supporting optional mixouts.

Key claim: Velaris preserves execution quality while reducing exploitable meta‑data leakage.
~50–150ms
RFQ quote latency
3+ legs
Typical hybrid routes
3 profiles
Privacy modes (fast/balanced/strong)

What makes Veil different

Hybrid routing

Combine firm RFQ fills with AMM depth based on a cost function that balances price, impact, and privacy profile.

Practical privacy

Session isolation, randomized windows, probabilistic batching, and optional mixouts reduce trivial linkage risks.

Receipts & SLOs

Clear receipts, guardrails, and observable SLOs for intent→receipt flow under normal operating conditions.

Routing at a glance

function planRoute(intent, profile):
  candidates = quoteRFQ(intent) + scanAMMs(intent)
  scored = []
  for c in candidates:
    cost = price(c) + slippage(c) + fees(c)
    privacy = leakagePenalty(c, profile)
    scored.append((c, cost + privacy))
  route = selectMinCost(scored, constraints=intent.bounds)
  return route

Principles

  • Execution first: competitive quotes + deterministic depth.
  • Privacy by construction: defaults reduce easy linkage.
  • Atomic settlement: clear receipts and error semantics.

Next

Explore Design Goals, the Architecture, or the Privacy Model.