FOR MARKET MAKERS

Solana infrastructure for market makers.

Market making is a latency game played from both sides: you must see the book move and pull or reprice your quotes before you get run over.

THE WORKLOAD

What decides the outcome.

A market maker posts two-sided quotes and continuously repositions them as the reference price and order book move. The risk is adverse selection - being filled on a stale quote because you saw the move late or could not cancel in time. That makes market making one of the most demanding workloads on Solana: it stresses the full stack at once - continuous streaming in, fast cancel/replace out.

  • You need a continuous, low-latency view of the market and oracle accounts - a polling loop is both too slow and too coarse to quote safely against.
  • Cancel latency is risk. The window between 'the price moved' and 'my stale quote is gone' is exactly where you get picked off, so cancel/replace must reach the leader fast.
  • Under congestion, unstaked transactions get deprioritized - so your cancels can fail precisely when volatility spikes and you most need them to land.

THE STACK

The path, mapped.

Yellowstone gRPC

Continuous, filtered streaming of the market and oracle accounts - the input a quoting loop needs, without polling lag.

Explore Yellowstone gRPC

Transaction sender

Fast cancel/replace straight to the leader, with staked-connection and Jito paths so your quotes update and your cancels land under load.

Explore Transaction sender

Dedicated nodes

Single-tenant bare metal so your tail latency and jitter are yours alone - no noisy neighbors between you and the book.

Explore Dedicated nodes

FAQ

Straight answers.

What infrastructure does a Solana market maker need?
A continuous low-latency market-data stream (Yellowstone gRPC) to see the book and oracle move, and a fast, reliable path to cancel and replace quotes (a direct-to-leader sender with staked connections). Dedicated, single-tenant capacity keeps tail latency deterministic. The failure mode to engineer against is adverse selection from slow cancels.
Why does cancel latency matter more than fill latency?
A missed fill is a lost opportunity; a stale fill is a realized loss. If the price moves and your quote is still resting because your cancel was slow, you get adversely selected. Minimizing the cancel path - stream in, cancel out, both co-located - is the core risk control.

Quoting safely on Solana is an infrastructure problem before it is a strategy problem. Let us map the stream-in, cancel-out path for your venue.