Best Hyperliquid RPC Providers: HyperEVM + HyperCore

Disclosure: CoinCodeCap may earn a commission if you sign up for Hyperliquid through links on this page. RPC provider links are unaffiliated. Risk warning: Crypto trading and infrastructure decisions involve risk — always load-test in your own environment before going to production.

How We Rank: CoinCodeCap evaluates Hyperliquid RPC providers on six factors: HyperCore + HyperEVM coverage, p95 latency under load, gRPC streaming support, free tier and pricing transparency, geographic node distribution, and uptime SLAs. Pricing is pulled directly from provider docs, and rate limits are cross-checked against the official Hyperliquid documentation at hyperliquid.gitbook.io.

Hyperliquid’s public RPC at rpc.hyperliquid.xyz/evm caps every IP at 100 requests per minute. That ceiling has been in place since August 9, 2025, and on a chain processing billions in daily perp volume with 0.2-second finality, it breaks any production application within minutes. Picking the right Hyperliquid RPC provider is the difference between a bot that scales and one that gets throttled at the worst possible moment.

Two architecture details to anchor before the comparison: HyperEVM (Chain ID 999) handles standard Solidity smart contracts with gas paid in HYPE — most providers cover this layer. HyperCore is the native trading engine; order placement, cancels, fills, and liquidations route through /info and /exchange endpoints, and HyperCore is gasless. Only a handful of providers stream HyperCore data via gRPC, which is what trading bots and market-making desks actually need.

This guide compares eight Hyperliquid RPC options — seven private providers plus the public endpoint — across pricing, gRPC streaming, regional latency, and uptime. The right pick depends on whether you’re shipping a HyperEVM dApp, a HyperCore-native trading bot, or both. If you don’t have an account yet, start with our Hyperliquid sign-up guide and our full Hyperliquid review for platform context.

What to Look For in a Hyperliquid RPC Provider

Six factors separate a Hyperliquid RPC provider that holds up in production from one that breaks at the first volatility spike. Run through these before committing to a paid plan:

• HyperCore + HyperEVM coverage: Most providers serve only HyperEVM. If your bot needs live order book data, fills, or funding rates, you need HyperCore access via REST, WebSocket, or gRPC. Confirm both before signing up.
• gRPC streaming for HyperCore: WebSocket polling can’t keep up with HFT-grade order flow. gRPC streams 70+ blocks per second with sub-second latency. As of early 2026, Dwellir, QuickNode (beta), and HypeRPC are the providers with serious gRPC offerings.
• Rate limits and burst handling: A 1,000 RPS ceiling means little if your provider throttles aggressively during volatility bursts. Benchmark p95 and p99 latencies under load — averages hide the tail.
• Geographic node distribution: Hyperliquid’s validators concentrate in Tokyo. Providers with JP edge servers (HypeRPC, Dwellir) cut round-trip time meaningfully for execution-sensitive workloads. EU regions are usually the cost-efficient default for everything else.
• Compliance certifications: SOC 2 Type II and ISO 27001 matter for institutional teams or anyone handling customer funds. Chainstack is the only provider with both certifications publicly listed as of this writing.
• Pricing model transparency: Per-request flat-rate pricing (Chainstack, Dwellir general plans) is predictable. Compute Unit models (Alchemy) require careful method-cost mapping. Tiered subscriptions (HypeRPC, QuickNode) are simplest if your traffic is steady.

How to Choose: A Decision Tree by Use Case

Skip the analysis paralysis. Match your use case to the right provider:

• Building a Hyperliquid trading bot or market maker? → HypeRPC (validator proximity in JP) or Dwellir (gRPC streaming + order book server). See our trading bot setup guide for the full stack.
• Need both HyperEVM contracts AND HyperCore data in one place? → Chainstack (single dashboard for both layers, with SOC 2 + ISO 27001 if compliance matters).
• Already on QuickNode for other chains? → QuickNode (unified billing, single console, HyperCore beta added January 2026).
• Just doing HyperEVM smart contract reads or DeFi integrations? → Alchemy (30M CUs free per month) or OnFinality (auto-scaling, multi-chain).
• Need cheap failover or multi-provider routing? → dRPC’s aggregator model with redundant routing across providers.
• Prototyping or scripting under 100 req/min? → Public RPC at rpc.hyperliquid.xyz/evm. Free, no signup, hard ceiling.

HypeRPC by Imperator

HypeRPC is the only Hyperliquid-only RPC provider on this list. It’s operated by Imperator.co, one of the largest multi-chain validators in Web3, and went live on March 20, 2025. The thesis is simple: validator-aligned infrastructure with edge servers in Tokyo (closest to Hyperliquid’s validator set) and the EU. Direct sentry peering eliminates network hops between your client and the validator set, and the published target is sub-500µs response time for latency-critical systems.

• ✅ Validator-aligned routing. Direct sentry peering in Tokyo, where Hyperliquid validators concentrate, cuts round-trip latency materially for execution-sensitive workloads.
• ✅ Full HyperCore coverage. REST + WebSocket for the Info API with 1:1 compatibility, L2 and L4 real-time order book streaming, and smart-filtered event streams.
• ✅ 99.99% uptime SLA on the Dedicated tier, with EU and JP regional failover and automatic routing to the fastest cluster.
• ⚠️ Higher entry price. Starts at $99/mo for the standard tier, which is steeper than Chainstack’s $49 starter or Alchemy’s free tier.
• ⚠️ Hyperliquid-only. No multi-chain support. If you also need Ethereum or Solana RPC, you’ll run two providers.
• 📌 Best for: Production trading bots, market-making desks, and HFT infrastructure where every millisecond from order placement to confirmation matters.

Chainstack

Chainstack is the most balanced choice for teams that need both HyperCore and HyperEVM access in a single dashboard. It’s also the only provider with both SOC 2 Type II and ISO 27001 certifications publicly documented — a real consideration for institutional teams. Chainstack added private Hyperliquid RPC endpoints via its GEN architecture, available on all paid plans, and exposes HyperEVM on two paths (/evm for hl-node-compliant responses, /nanoreth for system transactions included).

• ✅ Both layers, one dashboard. Private HyperEVM endpoints + HyperCore endpoints + a testnet HYPE faucet, archive access, and HyperEVM WebSockets — all from a single console.
• ✅ Free tier available. Useful for testing and small production loads before committing. Starts at $49/mo for the Growth tier with 20M requests included (~$2.45 per million requests).
• ✅ Compliance-ready. SOC 2 Type II and ISO 27001 certifications, with a 99.9% uptime SLA on dedicated nodes.
• ✅ Globally distributed nodes with predictable per-request flat pricing — every eth_call and eth_getLogs counts as 1 unit (archive 2x).
• ⚠️ Costs scale with volume. At very high request rates (10M+/month) the per-request model adds up. Compare against Dwellir’s $1,150/mo unlimited tier if you’re processing institutional volume.
• 📌 Best for: Teams building HyperEVM dApps that also need HyperCore data, fintech-adjacent projects with compliance requirements, and anyone wanting predictable pay-per-request billing.

Dwellir

Dwellir is the Swedish multi-chain provider that has gone deepest on Hyperliquid-specific tooling. Their offering goes well beyond standard RPC: dedicated gRPC streaming, an L2/L4 order book server with 100 levels per side (vs 20 on the public API) and individual order visibility, edge servers in Singapore and Tokyo for validator proximity, and co-located VPS instances where your code runs alongside the node. Their published benchmark shows a 24.1% median latency improvement over the public feed across 2,662 trades.

• ✅ HyperCore gRPC streaming. Dwellir is currently the only provider offering managed Hyperliquid gRPC endpoints with SLA-backed clusters — process 70+ blocks/sec with sub-second latency.
• ✅ L2/L4 order book server. Five times more depth than the public API (100 levels per side) plus L4 individual-order visibility including user addresses and timestamps.
• ✅ Co-located execution. Run single-tenant VPS instances alongside dedicated Hyperliquid nodes — eliminates network round-trips entirely. Useful for HFT and arbitrage.
• ✅ Transparent 1:1 pricing. One RPC response equals one credit, regardless of method complexity.
• ⚠️ No published compliance certifications. No public SOC 2 or ISO 27001 listing, which can be a blocker for institutional teams. Chainstack is the safer pick for compliance.
• ⚠️ No recurring free tier. Starter is a one-time $5; serious use starts at $49/mo, with the unlimited Hyperliquid dedicated node at $1,150/mo.
• 📌 Best for: HFT bots, market makers, and analytics teams building real-time data pipelines who need full L4 order book visibility and gRPC streaming.

QuickNode

QuickNode entered the Hyperliquid ecosystem in mid-2025 and launched a HyperCore public beta on January 20, 2026. The differentiator is the unified endpoint: QuickNode delivers seven HyperCore data streams (trades, orders, book updates, TWAP, events, blocks, writer actions) through gRPC, Streams, JSON-RPC, and WebSocket — all from a single endpoint that also serves HyperEVM. For multi-chain teams already on QuickNode, this is the path of least resistance.

• ✅ Unified endpoint. Single URL handles HyperEVM JSON-RPC, HyperCore gRPC, Streams, and WebSocket — no separate keys or dashboards.
• ✅ Globally distributed auto-scaling network routes traffic to the nearest region. Useful if your bot infrastructure is spread across multiple data centers.
• ✅ Established platform. If you already use QuickNode for Ethereum, Solana, or Base, adding Hyperliquid is a checkbox in the existing console.
• ⚠️ HyperCore is in beta. Production-readiness is real but evolving. Test thoroughly before committing critical workloads.
• ⚠️ Credit-based billing. Different methods consume different credit volumes, which can make cost projection less straightforward than Chainstack’s flat per-request model.
• 📌 Best for: Multi-chain trading desks, teams already on QuickNode for other networks, and developers who want one provider to rule them all.

Alchemy

Alchemy is the most generous free tier in this comparison and the most familiar to EVM developers. Up to 30 million Compute Units per month on the free tier — enough for many small production apps. Pay-as-you-go pricing starts at $5. The catch: Alchemy covers HyperEVM only. If your application doesn’t touch HyperCore order book data, this is the path of least resistance for Solidity developers used to Alchemy’s tooling.

• ✅ 30M CUs free per month. Easily the most permissive free tier — viable for small production apps without paying anything.
• ✅ Familiar EVM tooling. Same SDKs, same dashboards, same workflow as Ethereum or Polygon. Zero learning curve for EVM developers.
• ✅ Pay-as-you-go from $5. Unusually accessible entry point. Compute Unit pricing scales linearly with usage.
• ⚠️ HyperEVM only. No HyperCore order book or trading data. If you need fills, funding rates, or live order book access, this is not the right provider.
• ⚠️ Compute Unit accounting can make method-by-method cost projection tricky. Some calls (Debug API, Trace API) are not on the free tier at all.
• 📌 Best for: Solidity developers building HyperEVM dApps, hobbyist projects, and teams who already use Alchemy across other chains.

dRPC

dRPC takes a different approach: it’s a routing layer rather than a single-node provider, exposing a unified RPC endpoint backed by multiple underlying nodes. The pitch is redundancy and availability — if one upstream provider has issues, dRPC routes around them. For Hyperliquid specifically, the focus is HyperEVM. HyperCore-specific optimization is limited compared to specialized providers, but the failover characteristics make dRPC a strong secondary option in a multi-provider architecture.

• ✅ Routing and redundancy across multiple underlying providers improves availability if any single node has issues.
• ✅ Decentralized model — different from the centralized SaaS approach of QuickNode or Alchemy. Useful for teams who prefer non-single-vendor infrastructure.
• ✅ Free tier available with reasonable limits for testing and small workloads.
• ⚠️ HyperEVM only. No HyperCore order book streaming.
• ⚠️ Tail latency variability. Routing layers add a hop. Not the right pick if every millisecond matters.
• 📌 Best for: Failover routing, secondary RPC in a multi-provider design, and teams who care about decentralized infrastructure aesthetics.

OnFinality

OnFinality is a multi-chain RPC platform serving 130+ networks with auto-scaling infrastructure. For Hyperliquid, OnFinality offers HyperEVM access with no rate limits on the Ultimate tier and globally distributed endpoints. The positioning is straightforward: drop-in EVM RPC with enterprise-grade uptime monitoring, intended for teams that want auto-scaling without building it themselves.

• ✅ Auto-scaling infrastructure. Handles traffic spikes without manual resizing — useful during volatility events when bot traffic surges.
• ✅ 130+ networks supported. Multi-chain teams get unified billing and a single dashboard.
• ✅ Drop-in compatibility. Standard EVM RPC interface — no migration friction from Alchemy or QuickNode.
• ⚠️ HyperEVM only. No HyperCore streaming or trading data access.
• ⚠️ Less Hyperliquid-specific tooling than HypeRPC, Chainstack, or Dwellir. Treat it as a solid multi-chain default rather than a Hyperliquid specialist.
• 📌 Best for: Indexing-heavy applications, dashboards, multi-chain analytics platforms, and teams wanting low-friction auto-scaling.

Public RPC (rpc.hyperliquid.xyz/evm)

The public RPC is maintained by Hyperliquid’s core team and serves a single purpose well: free prototyping. It supports standard read-only JSON-RPC methods (eth_call, eth_blockNumber, eth_getTransactionByHash, eth_getLogs) and the HyperCore Info API endpoints for testing. The 100 requests-per-minute cap that took effect August 9, 2025 makes it unusable for any real production workload, but it’s perfect for scripts, learning, and load-testing your bot logic before paying for a private endpoint.

• ✅ Free, no signup required. Just point your client at rpc.hyperliquid.xyz/evm and you’re connected.
• ✅ Maintained by the core team. Same code path as official Hyperliquid infrastructure.
• ✅ Testnet faucet available for development against a full replica of mainnet.
• ⚠️ 100 requests per minute hard cap per IP. Any production app hits this in seconds during a volatility burst.
• ⚠️ Shared infrastructure means noisy-neighbor latency. Performance varies with overall network load.
• 📌 Best for: Prototypes, hackathons, light scripts, and load-testing your bot logic before provisioning private RPC.

Routing Verdict: Which Provider Wins for Each Use Case

The right Hyperliquid RPC provider depends on what you’re shipping. Here’s the verdict by builder profile:

• HFT bot or market maker: HypeRPC for execution-side latency in Tokyo, optionally paired with Dwellir for gRPC streaming and the L4 order book server.
• HyperEVM dApp + HyperCore data needs: Chainstack. Single dashboard, both layers, free tier to start, SOC 2 + ISO 27001 if you need it later.
• Existing QuickNode customer: QuickNode’s unified endpoint with HyperCore beta is the lowest-friction option.
• Solidity developer building HyperEVM only: Alchemy’s 30M-CU free tier ships fast, then graduate to Chainstack if you eventually need HyperCore.
• Multi-chain analytics or indexing: OnFinality for auto-scaling and 130+ network coverage.
• Backup or failover routing: dRPC as a secondary provider in a multi-provider architecture.
• Prototyping under 100 req/min: Public RPC. Don’t ship to production with it.

Two final operational notes. First: no private RPC provider can place orders on Hyperliquid through proprietary HyperCore methods — order placement and cancellation still route exclusively through Hyperliquid’s public API at api.hyperliquid.xyz. Private RPC accelerates reads and streams, not writes. Second: always benchmark p95 and p99 latency from your target region before committing. Average latency hides the tail behaviour that will break your bot during a volatility event.

Frequently Asked Questions

Can RPC providers place orders on Hyperliquid?

No. Order placement, cancels, and other proprietary HyperCore trading methods route exclusively through Hyperliquid’s public API at api.hyperliquid.xyz. Private RPC providers accelerate reads — querying state, streaming order book data, monitoring fills — but writes still go through the official endpoint. Your trading bot will use the official API for execution and a private RPC provider for everything else.

Why is the public Hyperliquid RPC rate-limited to 100 requests per minute?

The 100 rpm ceiling took effect August 9, 2025. The public endpoint is shared infrastructure, and as Hyperliquid grew into the largest perp DEX by volume, public-RPC load became unsustainable. Capping per-IP usage protects the shared resource for everyone running prototypes and lightweight scripts. Production workloads need a private RPC provider — the public endpoint was never designed to handle bot traffic at scale.

Do I need gRPC streaming for a Hyperliquid trading bot?

It depends on your strategy. For DCA bots, swing-trading bots, or any strategy that polls every few seconds, REST + WebSocket is usually fine. For HFT, market making, latency arbitrage, or any strategy where milliseconds matter, gRPC is the right call — it streams 70+ blocks per second with sub-second latency, which polling architectures can’t match. Dwellir, QuickNode (beta), and HypeRPC are the three providers offering serious gRPC for HyperCore as of early 2026.

The Hyperliquid public RPC’s 100 req/min cap means private infrastructure isn’t optional once you go to production — it’s foundational. Pick based on your actual workload: HypeRPC for trading-bot latency, Chainstack for both layers with compliance, Dwellir for gRPC streaming and HFT. Once your RPC is provisioned, the next step is funding your account and shipping your strategy. Our bridge and first trade guide walks through the USDC deposit and onboarding flow end to end.

Reviewed by Gaurav Agarwal, founder of CoinCodeCap. Gaurav has covered crypto exchanges, DeFi protocols, and trading platforms since 2018. Provider details verified against official documentation as of May 2026.

⚡ Bottom Line: The 100 req/min cap on Hyperliquid’s public RPC makes private infrastructure mandatory for production. HypeRPC delivers validator-aligned latency in Tokyo for trading bots ($99/mo). Chainstack covers both HyperCore and HyperEVM with SOC 2 + ISO 27001 compliance and a free starter tier. Dwellir is the only managed gRPC streaming provider with co-located execution ($1,150/mo unlimited). For HyperEVM-only work, Alchemy’s 30M-CU free tier is the path of least resistance. Public RPC is for prototyping, not production.

📋 Related Guides: Hyperliquid Trading Bot Setup | Hyperliquid Fees Explained | HLP Vaults & Yield
📊 Comparisons: Hyperliquid vs dYdX, GMX & Competitors
⬆️ Full Review: Hyperliquid Review — Is It Safe, Legit & Worth Switching?