Hyperliquid API Rate Limits and Address-Based Trading Constraints
Summary
This document explains Hyperliquid request limits for REST, WebSocket, explorer, and EVM JSON-RPC interfaces, alongside limits tied to individual trading addresses. It distinguishes per-IP request weights from address-based action limits, describes how batch size affects each, and outlines constraints on connections, subscriptions, messages, and simultaneous requests. It also notes that address limits depend on cumulative trading volume, with separate cancellation allowances and open-order caps.
The guidance is operational rather than a trading strategy. It recommends WebSockets for low-latency streaming and suggests an alternate data source for large historical block requests. Limits can vary in practical effect with response size, asset fee-tier weighting, congestion, and prior maker share, so systems should account for these rules when scheduling requests and managing orders. The document provides no performance evidence or detailed implementation example; its limits are platform-specific and may change over time.
Key ideas
- REST endpoints carry different request weights, and some weights rise with batch size or response length.
- WebSocket use is subject to caps on connections, subscriptions, users, messages, and inflight posts.
- Address-level action limits scale with cumulative trading volume and include a larger allowance for cancels.
- Each user has an open-order cap that can increase with volume, subject to a stated maximum.
- Batched orders count as one request for IP limits but as multiple requests for address limits.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.