Top 7 Ethereum RPC providers for RWA in 2026

Ethereum is the dominant settlement layer for tokenized real-world assets, holding the majority of tokenized real-world asset value (RWA.xyz tracks the live split by network). The sector crossed $26.4 billion total in March 2026, driven by institutional entries from BlackRock (BUIDL at $1.9B AUM), Ondo Finance ($2.7B across tokenized Treasuries and equity products), and Securitize ($3.3B in tokenized assets under administration) — all settled primarily on Ethereum.
What makes RPC infrastructure uniquely critical for RWA protocols is the compliance dimension: a missed event or a dropped NAV update does not just affect user experience — it can affect the legal integrity of the tokenized asset itself. ERC-3643 and ERC-1400 token standards require synchronous identity registry checks via eth_call on every transfer simulation, and eth_getLogs must be polled continuously without missing a single ComplianceCheck or Transfer event across the token contract and identity registry simultaneously.
This guide covers the seven RPC methods central to Ethereum RWA infrastructure, the infrastructure requirements that separate production-grade from development-grade endpoints, and a seven-provider comparison evaluated specifically through the lens of regulated asset workloads.
Quick verdict: best Ethereum RPC providers for RWA in 2026
Chainstack is our top pick for compliance-critical RWA on Ethereum (an editorial judgment, not a benchmark result), based on dual SOC 2 Type II and ISO 27001 certification, method-agnostic pricing, archive and debug/trace on paid plans, Dedicated Nodes and a contractual SLA on Enterprise. Chainstack publishes this page, runs the benchmark cited below and is one of the 7 listed providers, so see how we compared and the benchmark limits before relying on the ranking. In that benchmark Chainstack had the lowest overall P50, P95 and P99 of the three listed providers measured, but its US P95 trailed Alchemy and Quicknode. Last updated October 29, 2026. The best fit by need:
- Best for compliance-critical RWA with institutional due diligence: Chainstack (dual SOC 2 Type II + ISO 27001, method-agnostic pricing, archive on paid plans, Dedicated Nodes).
- Best for managed failover across many upstream providers: RouteMesh (routing layer; no certification of its own).
- Best for flat per-response billing and a dedicated archive cluster: Dwellir (trace/debug on paid plans; no public SOC 2 or ISO 27001 found).
- Best for issuers already evaluating custody or staking vendors: Blockdaemon (SOC 2 + ISO 27001 stack, quote-based dedicated nodes).
- Best for ConsenSys-ecosystem teams that want archive on every plan: Infura (no public SOC 2 or ISO 27001 listed on its pricing page).
- Best for developer-centric early-stage RWA dApps: Alchemy (30M CU/month free; fastest US P95 of the three providers measured in the October 2026 benchmark, 80 ms).
- Best for teams that need ISO 27001 alongside SOC 2 from a direct competitor: Quicknode (fastest Germany P95 of the three providers measured, 40 ms).
Provider comparison: 7 Ethereum RPC providers for RWA (2026)
The table below summarizes public positioning as of October 2026. The rank is the editorial order used in this article (see how we compared), not a benchmark result. The last column shows the overall P95 from Chainstack’s October 2026 Ethereum benchmark, which measured only three of the seven providers (lower is better). Ankr and dRPC, both meaningful Ethereum providers in general, are left out of this specific comparison — neither offers a public audit report suited to RWA vendor due diligence (Ankr shares its SOC 2 Type II report only under NDA), which is the deciding criterion for this use case rather than raw chain coverage. Dwellir is included for its dedicated archive clusters and SLA terms, with the caveat that Dwellir also has no public attestation.
| Rank | Provider | Best for | Pricing model | Free tier | Dedicated nodes | Archive & trace | Why it matters for RWA | Overall P95 in Oct 2026 benchmark (Chainstack-run) |
|---|---|---|---|---|---|---|---|---|
| 1 | Chainstack | Compliance-critical RWA: dual SOC 2 Type II + ISO 27001, method-agnostic pricing, archive on paid plans | Method-agnostic (1 RU standard, 2 RU archive/trace) | 3M RU/month, 25 RPS, no card required | Yes (Pro plan and up) | Yes, paid plans (Growth and up) | Dual SOC 2 Type II + ISO 27001, flat pricing, and MEV protection by default reduce cost uncertainty and vendor due-diligence effort for regulated workloads | 78 ms |
| 2 | RouteMesh | Managed failover across many upstream providers | Per-request, tiered by routing class ($3–6/1M requests) | Free public community endpoint | No (routing layer) | Depends on the upstream node it routes to | Automatic failover across 15+ underlying providers reduces the single-endpoint downtime risk RWA compliance polling can’t tolerate | Not measured |
| 3 | Dwellir | Flat per-response billing and dedicated archive clusters | Method-agnostic (1 credit per response) | 20 responses/s, 100,000 responses/day | Yes, dedicated clusters listed from $2,400/mo per chain (Ethereum archive cluster price not verified) | Trace/debug on paid plans; full archive state on dedicated clusters | Flat per-response billing and a 99.99% uptime SLA claimed on Dwellir’s marketing pages (contract terms not verified), but no SOC 2 or ISO 27001 attestation found publicly | Not measured |
| 4 | Blockdaemon | Issuers already evaluating custody or staking vendors | Free API tier and self-serve Starter/Growth plans; dedicated nodes and Enterprise by quote | Yes (5 RPS API tier) | Yes, node clusters with automated failover | Not publicly detailed for Ethereum — confirm per contract | SOC 2 + ISO 27001 institutional stack, serving 400+ institutions as stated by Blockdaemon | Not measured |
| 5 | Infura | ConsenSys-ecosystem teams, archive on every plan | Credit-based, method-weighted | Limited credits/day | No dedicated node product listed | Archive on all plans; debug/trace from Developer | Ethereum-native with ConsenSys institutional credentials; no SOC 2 or ISO 27001 listed on its pricing page | Not measured |
| 6 | Alchemy | Developer-centric early-stage RWA dApps | CU-weighted (26 CU/eth_call) |
30M CUs/month | No (private endpoints) | Archive on all plans; Debug/Trace on pay-as-you-go and above | Strong developer tooling; CU weighting inflates cost for compliance-heavy eth_call workloads |
103 ms |
| 7 | Quicknode | Teams that need ISO 27001 alongside SOC 2 from a direct competitor | Credit-weighted by method | 1-month trial only | Dedicated clusters | Archive and trace/debug on paid plans | SOC 2 Type II + ISO 27001 (reports on request on the enterprise tier, per Quicknode’s pricing page); dedicated clusters | 92 ms |
How we compared Ethereum RPC providers for RWA
Disclosure: this comparison is published by Chainstack. Chainstack is one of the 7 listed providers and is ranked first, so read the ranking as a vendor perspective and check the figures against the official pricing pages linked in each provider section.
Criteria. Each provider is compared on the same RWA-relevant attributes: archive and debug/trace access, dedicated infrastructure, published certifications (SOC 2 Type II and ISO 27001), contractual SLA terms, pricing model and free tier. The order is editorial judgment for regulated asset workloads. It is not a benchmark result. The latency benchmark below covers only 3 of the 7 listed providers (Chainstack, Alchemy and Quicknode), so it informs the latency discussion for those three and does not determine the order.
Sources. Prices, limits and certifications come from each provider’s public pricing, documentation and security pages, as summarized on this page as of October 2026 unless a section says otherwise. Latency figures for Chainstack, Alchemy and Quicknode come from the Chainstack Labs Research Team’s October 2026 Ethereum benchmark. Chainstack’s platform uptime is shown as 99.99%+ platform uptime (status.chainstack.com, Ethereum Mainnet, 99.99%+ over the 90 days to October 29, 2026). That figure is platform uptime from the status page, which covers the Mainnet component of each chain (testnet and devnet components are tracked separately) and Chainstack only; it is separate from the benchmark’s probe success rate and from contractual SLAs, and it is not a guarantee.
Not covered. Ankr and dRPC are left out of the 7 for the reason given above the comparison table. dRPC appears in the benchmark table because the report measured it, but it is not one of the 7 listed providers. Other Ethereum RPC providers may also exist beyond these 7. RouteMesh, Dwellir, Blockdaemon and Infura were not measured by the benchmark.
Latency data. See the Ethereum latency benchmark section for the figures, the method and what the benchmark does not show. For your own workload, measure from the region where your application runs; see how to measure RPC latency (P50, P95, P99).
Already using Chainstack? Jump to the Ethereum tooling docs or deploy your endpoint in minutes.
RWA on Ethereum: RPC requirements
Real-world asset protocols are not typical DeFi workloads. They combine the data throughput of DeFi indexers with the uptime requirements of payment infrastructure and the compliance obligations of regulated financial services. Understanding what this means for your RPC selection starts with the specific methods these protocols call — and how they call them.
Latency requirements
RWA infrastructure is not latency-critical in the way HFT or MEV is, but it is consistency-critical. A tokenized treasury fund polling for oracle updates every 60 seconds can tolerate 200ms response times — but cannot tolerate a 30-second gap where the endpoint is unavailable or rate-limited. Transfer compliance checks via eth_call must complete before a transaction is broadcast, so p99 latency matters as much as median for interactive flows.
The practical threshold: p95 below 500ms for eth_call, under 2 seconds for paginated eth_getLogs across a 5,000-block range. Archive-dependent lookups for audit trail reconstruction are latency-tolerant but must be reliable — a failed debug_traceTransaction during a regulatory review window is not recoverable with a retry.
Throughput requirements
A typical RWA protocol in production generates:
- Continuous
eth_callload — 1 to 5 eligibility checks per user action, plus background polling against the ONCHAINID registry - Periodic
eth_getLogsbursts during settlement and distribution windows - Steady WebSocket subscription traffic for real-time Transfer and ComplianceCheck event monitoring
- Occasional
debug_traceTransactioncalls triggered by compliance workflow or audit reconstruction
This translates to 50–200 RPS on active protocols with burst patterns during settlement cycles. Public shared endpoints with 10–25 RPS caps are inadequate for anything beyond prototyping.
Key RPC methods for Ethereum RWA
These seven methods form the core RPC surface for ERC-3643 and ERC-1400 protocols. Every provider in this comparison is evaluated against their ability to handle this method mix reliably at production scale.
| Method | What it does for RWA | Infrastructure concerns |
|---|---|---|
eth_call |
Checks investor eligibility against ONCHAINID / identity registry before every transfer | Runs pre-broadcast on every transfer attempt; absorbs continuous load — method weighting is expensive on CU platforms |
eth_getLogs |
Indexes Transfer, ComplianceCheck, and OracleUpdate events across token + registry contracts | Public endpoints throttle this aggressively; 5,000-block range limit requires careful pagination and retry logic |
eth_getStorageAt |
Retrieves historical contract state for proof-of-reserve and audit reconstruction | Requires archive node access — not available on full node endpoints |
debug_traceTransaction |
Provides transaction-level execution trace for regulatory audit trails | Debug namespace not available on all provider tiers; requires dedicated or archive-enabled plans |
eth_sendRawTransaction |
Broadcasts compliant token transfers after eligibility verification | Standard on all providers; failover and broadcast reliability matter more than raw speed |
eth_getTransactionReceipt |
Confirms settlement finality and extracts event logs from executed transfers | Latency and reliability critical for downstream settlement logic triggered on confirmation |
eth_subscribe("logs") |
Real-time streaming of contract events via WebSocket | Requires WebSocket endpoint with reconnect support; shared endpoints often cap concurrent subscriptions |
For the full reference of Ethereum RPC methods available through Chainstack, see the Ethereum API reference: JSON-RPC and Beacon quickstart.
Infrastructure requirements
Archive node access is mandatory, not optional. Any RWA protocol performing historical state reconstruction for audits, historical NAV lookups, or regulatory reporting needs archive data. Several providers gate archive access behind premium tiers or charge separately — this must be confirmed at contract time, not at incident time.
Dedicated endpoints for data residency. Many RWA protocols serving regulated investors face data residency requirements. Public shared endpoints route through infrastructure with no data residency guarantees. Dedicated Nodes provide isolated infrastructure with known data flow paths — a prerequisite for jurisdictional compliance in several EU and APAC markets.
WebSocket stability for event subscriptions. eth_subscribe("logs") subscriptions must survive network hiccups without dropping events. Production implementations require WebSocket endpoints with automatic reconnection and a fallback eth_getLogs polling strategy. Providers with instability at the WebSocket layer create compliance gaps — a dropped subscription during a settlement window means missed events with no automatic recovery.
Rate limit predictability over cost minimization. The cost structure for RWA is dominated by eth_call and eth_getLogs. Providers that weight these methods heavily in their CU or credit calculations impose a significant cost premium and, more importantly, unpredictable rate consumption that makes burst capacity planning difficult. Method-agnostic pricing removes this variable from compliance infrastructure planning.
SOC 2 Type II and ISO 27001 certification for vendor due diligence. Regulated asset issuers performing vendor due diligence on infrastructure providers — a standard requirement for funds and licensed platforms — need documented, audit-backed security certifications covering both operational controls and information security management. Self-reported compliance postures are not acceptable for institutional review.
Chainstack for RWA on Ethereum
Chainstack addresses the critical requirements for production RWA infrastructure simultaneously: archive and debug/trace access on every paid plan, dual SOC 2 Type II (December 2025) and ISO 27001 (July 2026) certification, and method-agnostic pricing where 1 RPC call = 1 request unit for standard reads.
The pricing model is particularly relevant for RWA workloads. Standard eth_call and eth_getBlockByNumber calls cost 1 RU; archive reads and debug_traceTransaction/trace_* calls cost 2 RU regardless of how far back they reach (see current multipliers on request units). eth_getLogs follows the same block-age rule — 1 RU within roughly 127 blocks of the tip, 2 RU for historical ranges. That flat structure removes the cost uncertainty that makes RWA infrastructure budgeting unpredictable on CU-weighted platforms: a protocol running 500,000 compliance checks per month against recent state spends the same RU budget as 500,000 standard block queries.
For regulated RWA deployments, Chainstack provides:
- Dedicated Nodes from the Pro plan — isolated infrastructure with known data paths for data residency compliance
- Unlimited Node add-on — flat-fee RPS tiers starting at $149/month for 25 RPS up to $3,199/month for 500 RPS, with zero per-request billing, suited to continuous background indexing workloads
- A contractual SLA (99.9% quarterly uptime commitment and service credits) on Enterprise; Chainstack platform uptime is shown on the public status page (see how we compared)
- MEV protection shipped by default on Ethereum Global Nodes, routing signed transactions to a private block-builder network instead of the public mempool — relevant when an RWA token’s secondary-market liquidity touches AMM pools exposed to front-running
- Dual SOC 2 Type II and ISO 27001 certification, covering security, availability, confidentiality, and a full information security management system
Current plan pricing: Developer free (3M RU/month, 25 RPS); Growth $49/month (20M RU, 250 RPS); Pro $199/month (80M RU, 400 RPS); Business $499/month (200M RU, 600 RPS, includes role-based access control); Enterprise from $990/month (400M+ RU, custom throughput). Annual billing saves up to 16%; see current figures on pricing since tiers are revised over time.
For regulated asset platforms specifically, see blockchain infrastructure for RWA, blockchain infrastructure built for fintech, and stablecoin infrastructure. For institutional deployments requiring contractual SLA and dedicated account management, see enterprise.
Prior Chainstack research on RWA-specific RPC requirements, including ERC-3643 infrastructure patterns and oracle monitoring workflows, is covered in depth at RPC infrastructure for RWA: EVM node requirements.
Code example: indexing ERC-3643 compliance events
The most common RWA infrastructure task is continuously indexing Transfer and ComplianceCheck events across the token contract and ONCHAINID identity registry. Both contracts must be watched together — an indexer watching only the token contract can miss registry updates that affect transfer eligibility:
import { ethers } from "ethers";
const provider = new ethers.JsonRpcProvider("YOUR_CHAINSTACK_ENDPOINT");
// ERC-3643 Transfer and ComplianceCheck event signatures
const TRANSFER_TOPIC = ethers.id("Transfer(address,address,uint256)");
const COMPLIANCE_TOPIC = ethers.id("ComplianceAdded(address,address)");
const RWA_TOKEN_ADDRESS = "YOUR_TOKEN_CONTRACT";
const IDENTITY_REGISTRY = "YOUR_IDENTITY_REGISTRY";
async function indexComplianceEvents(fromBlock, toBlock) {
const logs = await provider.getLogs({
address: [RWA_TOKEN_ADDRESS, IDENTITY_REGISTRY],
topics: [[TRANSFER_TOPIC, COMPLIANCE_TOPIC]],
fromBlock,
toBlock,
});
for (const log of logs) {
console.log(`Block ${log.blockNumber}: ${log.topics[0].slice(0, 10)} from ${log.address}`);
}
}
indexComplianceEvents("earliest", "latest");
The Ethereum tooling documentation includes full SDK examples for Hardhat, Foundry, ethers.js, viem, and web3.py.
Ethereum RPC rate limits and free tiers compared
Rate limits and billing units differ between providers, so the same number of calls can cost different amounts. This table uses only the figures given in this article; “Not stated here” means this article gives no figure.
| Provider | Free tier | Free-tier rate limit | Paid tiers or flat-rate throughput | Billing unit |
|---|---|---|---|---|
| Chainstack | Permanent Developer plan, 3M RU/month, no card required | 25 RPS | Growth $49/month (20M RU, 250 RPS); Pro $199/month (80M RU, 400 RPS); Business $499/month (200M RU, 600 RPS); Enterprise from $990/month (400M+ RU, custom throughput); Unlimited Node add-on 25 RPS at $149/month to 500 RPS at $3,199/month | Request Units: 1 RU standard, 2 RU archive and trace |
| RouteMesh | Free public community endpoint | Not stated here | $3 per million requests standard, $6 per million premium routing | Per request, tiered by routing class |
| Dwellir | Free plan: 100,000 responses/day | 20 responses/second | Developer $49/month (25M responses); Growth $299/month (150M); Scale $999/month (500M, 5,000 responses/second); Unlimited Nodes from $119/month; dedicated clusters from $2,400/month per chain (Ethereum archive cluster price not verified) | 1 credit per response (method-agnostic) |
| Blockdaemon | Free API tier | 5 RPS | Self-serve Starter and Growth plans (prices not stated here); dedicated nodes and Enterprise by quote | Not stated here |
| Infura | Free Core plan, limited credits/day (amount not stated here) | Not stated here | Debug and trace from the Developer plan (price not stated here) | Credits (method-weighted) |
| Alchemy | 30M CU/month, no time limit | Not stated here | Debug/Trace on pay-as-you-go and above (prices not stated here); private endpoints instead of dedicated nodes | Compute units (26 CU per eth_call) |
| Quicknode | One-month trial, 10M credits (no permanent free plan) | 15 RPS | Build $49/month (50 RPS); Accelerate $249/month; Scale $499/month; Business $999/month; dedicated clusters | Credits (method-weighted) |
Chainstack alternatives for RWA on Ethereum
If Chainstack is not the right fit, these are the head-to-head differences between Chainstack and each alternative, using the facts in the sections below:
- Chainstack vs RouteMesh: RouteMesh adds automatic failover across 15+ upstream providers and covers 1,000+ EVM networks, but operates no dedicated Ethereum infrastructure of its own and publishes no certification; Chainstack runs its own nodes and holds SOC 2 Type II and ISO 27001.
- Chainstack vs Dwellir: Dwellir bills one credit per response, lists dedicated clusters from $2,400/month per chain and claims a 99.99% uptime SLA on its marketing pages (contract terms not verified), but no SOC 2 or ISO 27001 attestation was found; Chainstack publishes both certifications and offers a contractual SLA on Enterprise.
- Chainstack vs Blockdaemon: Blockdaemon bundles custody, staking and tokenization services with a SOC 2 and ISO 27001 stack for 400+ institutions, but dedicated nodes and Enterprise are quote-based and Ethereum archive depth is not publicly detailed; Chainstack publishes its plan prices.
- Chainstack vs Infura: Infura has deep ConsenSys ecosystem ties and archive data on every plan, but lists no SOC 2 or ISO 27001 attestation on its pricing page and has no dedicated node product; Chainstack publishes both certifications.
- Chainstack vs Alchemy: Alchemy has a larger free tier (30M CU/month), Notify webhooks, SOC 2 Type II and a faster US P95 (80 ms against 143 ms) in the October 2026 benchmark, but weights
eth_callat 26 CU, offers no dedicated nodes and publishes no ISO 27001. Chainstack had the lower overall P50, P95 and P99 in that benchmark. - Chainstack vs Quicknode: Quicknode lists SOC 2 and ISO 27001 reports on request on its enterprise tier, offers dedicated clusters and had faster Germany (40 ms against 47 ms) and US (98 ms against 143 ms) P95 in the October 2026 benchmark, but has no permanent free plan and uses method-weighted credits. Chainstack had the lower overall P50, P95 and P99.
In-depth reviews: 7 Ethereum RPC providers for RWA (2026)
1. Chainstack: best for compliance-critical RWA with institutional due diligence

Verdict: Dual SOC 2 Type II and ISO 27001, method-agnostic pricing, archive and debug/trace on paid plans and a contractual SLA on Enterprise. Trade-off: the free tier’s 25 RPS ceiling is too low for production polling, the Enterprise SLA needs the $990/month plan, and its US P95 (143 ms) trailed Alchemy and Quicknode in its own October 2026 benchmark. Official pricing.
Chainstack’s position for RWA infrastructure combines dual SOC 2 Type II and ISO 27001 certification, method-agnostic pricing, and Dedicated Nodes with contractual SLA — the requirements that appear most consistently on institutional vendor due diligence checklists. For teams building ERC-3643 or ERC-1400 protocols, the absence of method weighting means 10,000 eth_call compliance checks against recent state cost exactly the same as 10,000 eth_getBlockByNumber calls, removing cost modeling complexity from compliance infrastructure planning.
Global Nodes provide geo-balanced shared infrastructure across multiple regions, appropriate for teams at early to mid-scale. The Pro plan and above unlock Dedicated Nodes for infrastructure isolation and predictable throughput. The Unlimited Node add-on converts billing entirely to a flat monthly fee across 25–500 RPS tiers — in our assessment a good model for RWA protocols running continuous background compliance polling and event indexing that generate high but predictable request volume. Archive and debug/trace access are included on every paid plan (Growth and up), billed at the 2 RU archive rate rather than gated behind a separate add-on.
Limitations: The free tier caps at 25 RPS, insufficient for production-scale continuous compliance polling. Enterprise SLA and dedicated account management require the $990/month Enterprise plan.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Excellent — method-agnostic pricing, archive on paid plans, dual certification, and Dedicated Nodes in our assessment make this a strong fit for compliance-critical workloads
- Archive and audit trail access: Excellent —
debug_traceTransactionand full archive available on paid plans (Growth and up) at the 2 RU rate - Enterprise / regulated deployments: Excellent — contractual SLA, isolated Dedicated Nodes, and dual audit reports meet institutional due diligence requirements that a single certification often doesn’t satisfy on its own
2. RouteMesh: best for managed failover across upstream providers

Verdict: Routing layer with automatic failover across 15+ providers. Trade-off: no dedicated Ethereum infrastructure or certification of its own, and archive and trace depend on the backend that answers. Pricing: $3 per million requests standard and $6 per million premium routing, as stated on this page; no official pricing link has been verified for this page.
RouteMesh is an intelligent RPC routing layer covering 1,000+ EVM networks, aggregating 15+ underlying providers behind a single endpoint with real-time health scoring and automatic failover between them. Pricing is $3 per million requests standard and $6 per million for premium routing. Customers include LI.FI (routing 97% of its EVM RPC calls through RouteMesh across 60+ chains), Vaults.fyi, and DAMM Capital.
For RWA infrastructure specifically, the value is availability rather than raw speed: the introduction section above notes that a tokenized treasury protocol can tolerate 200ms response times but not a 30-second endpoint outage. RouteMesh’s routing across multiple upstream providers means a single provider incident doesn’t take the endpoint down — traffic reroutes automatically to a healthy backend. RouteMesh is a routing and monitoring layer rather than a node operator, so archive depth, trace availability, and compliance certifications depend on which underlying provider a given request lands on, not on RouteMesh itself.
Limitations: RouteMesh doesn’t operate its own dedicated Ethereum infrastructure or publish its own compliance certification — teams needing a single attestable vendor for audit purposes will still need to evaluate the underlying providers it routes to. Archive and trace support varies by which backend serves a given request.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Good — the failover model directly reduces gap risk during compliance polling, though it adds a routing hop between the app and the node that actually holds compliance certification
- Archive and audit trail access: Moderate — depends entirely on the underlying provider a request routes to at that moment
- Enterprise / regulated deployments: Limited — no published SOC 2 or ISO 27001 of its own; institutional buyers will need to pair it with a certified upstream provider for the compliance story
3. Dwellir: best for flat per-response billing and dedicated archive clusters

Verdict: One credit per response, trace/debug on paid plans and a dedicated archive cluster; a 99.99% uptime SLA is claimed on its marketing pages (contract terms not verified). Trade-off: no public SOC 2 or ISO 27001 attestation found. Official pricing.
Dwellir is a Swedish RPC provider with a long Polkadot track record that now serves 150+ networks, Ethereum among them. For RWA teams the relevant pieces are a managed API over HTTPS and WebSocket with debug and trace methods on paid plans, eth_getLogs over up to 10,000 blocks from the Growth plan, and dedicated archive clusters with full archive state, no rate limits, and a 99.99% uptime SLA claimed on its marketing pages (contract terms not verified).
Billing is method-agnostic at one credit per response, which suits eth_call-heavy eligibility checks: Free covers 20 responses/second and 100,000 responses per day, Developer is $49/month (25M responses), Growth $299/month (150M), and Scale $999/month (500M, 5,000 responses/second). Unlimited Nodes start at $119/month flat, and dedicated clusters start at $2,400/month per chain (the Ethereum archive cluster price is not verified).
Limitations: We found no SOC 2 or ISO 27001 attestation on Dwellir’s public pricing and dedicated-cluster pages, which makes it a weaker fit than Chainstack, Quicknode, or Blockdaemon wherever vendor due diligence requires audit reports. Teams that still want it should request compliance documents before committing.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Good — flat per-response billing keeps continuous eligibility checks predictable; compliance evidence has to come from elsewhere
- Archive and audit trail access: Strong — trace/debug on paid plans and full archive state on dedicated clusters
- Enterprise / regulated deployments: Limited — a 99.99% uptime SLA is claimed on its marketing pages (contract terms not verified), but no published SOC 2 or ISO 27001
4. Blockdaemon: best for issuers already evaluating custody or staking vendors

Verdict: SOC 2 and ISO 27001 stack bundled with custody, staking and tokenization. Trade-off: dedicated nodes and Enterprise are quote-based and Ethereum archive depth is not publicly detailed. Official site (pricing on request).
Blockdaemon is institutional blockchain infrastructure built for custody, staking, tokenization, and node operation in one bundle, serving 400+ institutions including banks, asset managers, and exchanges across $110B+ in secured assets. Ethereum RPC access runs through production node clusters with automated scaling and failover, positioned explicitly for regulated buyers rather than individual developers.
For RWA specifically, Blockdaemon’s certification stack — SOC 2 and ISO 27001, alongside OFAC/sanctions-list compliance — gives institutional issuers the audit-backed posture their procurement teams expect. The tradeoff is procurement friction: its docs list a free API tier (5 RPS) and self-serve Starter and Growth plans, but dedicated nodes and Enterprise are quote-based, so a production RWA deployment typically ends in a sales conversation.
Limitations: Dedicated nodes and Enterprise terms are quote-based, so a production deployment requires a sales cycle beyond the entry API tiers. Archive depth and trace method availability for Ethereum specifically aren’t detailed publicly — confirm directly for compliance-heavy audit workloads.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Strong — the compliance and custody bundle matches what a regulated issuer’s procurement team is already looking for, once the sales process clears
- Archive and audit trail access: Moderate — dedicated node clusters support it, but per-chain archive depth for Ethereum isn’t published
- Enterprise / regulated deployments: Excellent — this is Blockdaemon’s core buyer; 400+ institutional customers and a dual-certification stack built specifically for this audience
5. Infura: best for ConsenSys-ecosystem teams that want archive on every plan

Verdict: Archive data on every plan and ConsenSys institutional credentials. Trade-off: no dedicated node product, no public compliance attestation listed and credit-based, method-weighted pricing. Official pricing.
Infura is one of the longest-established Ethereum-native providers, operating as ConsenSys‘ infrastructure layer and the default RPC for MetaMask. For teams with existing ConsenSys ecosystem relationships, or building wallets and dApps targeting MetaMask’s user base, Infura offers depth of Ethereum protocol integration and institutional familiarity that carries weight in vendor selection discussions with regulated platforms.
Full archive data is listed on every plan, including the free Core tier, while the debug/trace API starts on the Developer plan. eth_getLogs with deep historical range and debug_traceTransaction both depend on archive-enabled endpoints. Credit-based pricing is method-weighted, with eth_getLogs costing substantially more credits than simpler reads. Infura’s pricing page lists no SOC 2 or ISO 27001 attestation of its own; teams should confirm with ConsenSys which reports cover Infura before citing either in a procurement document. The Enterprise plan adds auto-scaling and an enhanced support SLA.
Limitations: There is no dedicated node product, and no public compliance attestation, which adds procurement friction for regulated RWA protocols. Standard plans do not include a contractual SLA.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Strong — deep Ethereum protocol expertise; no public compliance attestation to hand to a procurement team
- Archive and audit trail access: Good — archive data on every plan; debug/trace from the Developer plan
- Enterprise / regulated deployments: Moderate — Enterprise plan offers an enhanced support SLA and ConsenSys institutional credentials, but no public attestation or dedicated nodes
6. Alchemy: best for developer-centric early-stage RWA dApps

Verdict: 30M CU/month free tier, webhooks and published SOC 2 Type II; fastest US P95 (80 ms) of the three providers measured in the October 2026 benchmark. Trade-off: 26 CU per eth_call, no dedicated nodes and no published ISO 27001. Official pricing.
In our assessment, Alchemy is a strong option for Ethereum developer tooling, with a mature SDK ecosystem, Notify webhooks for event delivery, and a free tier at 30M CUs per month with no time limit. The Transact API adds reliability guarantees for transaction broadcast, and the webhook infrastructure reduces reliance on polling eth_getLogs for event-driven workflows.
The CU structure needs careful evaluation for production RWA workloads: eth_call currently costs 26 CU per request. An RWA protocol running 100,000 compliance checks daily consumes 2.6M CUs per day — roughly 9% of the monthly free tier every day, so the 30M-CU allowance lasts under 12 days at that rate. Alchemy publishes SOC 2 Type II certification but does not publish ISO 27001 — a distinction worth checking directly if a compliance team specifically requires both. Alchemy does not offer dedicated nodes in the traditional sense; infrastructure isolation runs through private endpoints, which don’t provide guaranteed throughput under load the way a dedicated cluster does.
Limitations: CU weighting makes Alchemy expensive at scale for eth_call-heavy compliance workloads. No dedicated nodes means no guaranteed throughput isolation for production RWA protocols during busy settlement windows.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Good — archive access on paid tiers; CU weighting creates cost unpredictability at scale
- Archive and audit trail access: Good — archive on all plans; Debug/Trace on pay-as-you-go and above
- Enterprise / regulated deployments: Moderate — SOC 2 Type II only; no dedicated nodes and no standard contractual SLA below Enterprise tier
7. Quicknode: best for teams that need ISO 27001 alongside SOC 2 from a direct competitor

Verdict: SOC 2 Type II and ISO 27001 reports on request on the enterprise tier and dedicated clusters; fastest Germany P95 (40 ms) of the three providers measured in the October 2026 benchmark. Trade-off: no permanent free plan, method-weighted credits and a higher list price per unit of throughput than Chainstack. Official pricing.
Quicknode publishes SOC 2 Type II and ISO 27001 (reports on request on its enterprise tier, according to its pricing page). For RWA protocols subject to regulatory audit, where vendor attestation must be presented to compliance officers, that certification stack is straightforward to cite. Archive access and trace/debug methods are available on paid plans (trace and debug start on the Build plan), and dedicated clusters are available for teams needing isolated throughput.
Current plans run from Build ($49/month) through Accelerate ($249/month), Scale ($499/month), and Business ($999/month) up to custom Business+ and Enterprise tiers. Build costs the same $49/month as Chainstack’s Growth plan but tops out at 50 RPS against Growth’s 250 RPS. There’s no permanent free plan; new accounts get a one-month free trial (10M credits, 15 RPS) before it converts to a paid subscription. Credit-weighted pricing for eth_call and eth_getLogs increases effective cost for compliance-heavy workloads relative to method-agnostic pricing — teams evaluating Quicknode for continuous compliance polling should model their specific method mix against a cost comparison before committing to a tier.
Limitations: No permanent free tier — only a one-month trial. Credit-weighted pricing for eth_call-heavy RWA compliance workloads means the effective cost per check climbs faster than the sticker price suggests.
Fit by workload:
- RWA compliance (ERC-3643/ERC-1400): Excellent — SOC 2 Type II and ISO 27001 published; archive and trace/debug on paid plans, trace/debug from the Build plan
- Archive and audit trail access: Excellent — archive and debug/trace on paid plans
- Enterprise / regulated deployments: Excellent — SOC 2 Type II + ISO 27001 provides broad vendor attestation for institutional due diligence, at a higher list price per unit of throughput than Chainstack
Ethereum RPC latency benchmark (October 2026): Chainstack, Alchemy and Quicknode
Disclosure: Chainstack operates the monitoring service, the public Grafana dashboard and the scoring formula behind this benchmark, and is one of the four providers measured. The figures below come from the Ethereum RPC provider benchmark (Research Note No. 050, published October 8, 2026) by the Chainstack Labs Research Team. Live data is on Chainstack Compare and the Chainstack performance dashboard. Compare labels the success-rate column “availability”; here it is called probe success rate because it is not platform uptime.
Method in brief: a 7-day window ending October 8, 2026; probes every 1 to 3 minutes from three cloud regions (Germany, US and Singapore) using a Python service on Vercel serverless functions; each provider on its first paid plan; a request counts as failed if it takes longer than 55 seconds. The score is response time divided by success rate cubed per region (lower is better), averaged equally across the three regions. The latency term is the mean of each method’s P95 and excludes eth_subscribe. The collection code is open source (compare-dashboard-functions).
| Provider | Listed in this article | Score (lower is better) | Probe success rate | P50 (ms) | P95 (ms) | P99 (ms) | Germany P95 (ms) | US P95 (ms) | Singapore P95 (ms) |
|---|---|---|---|---|---|---|---|---|---|
| Chainstack | Yes | 0.079 | 99.98% | 36 | 78 | 117 | 47 | 143 | 42 |
| Quicknode | Yes | 0.097 | 99.98% | 40 | 92 | 142 | 40 | 98 | 138 |
| Alchemy | Yes | 0.114 | 99.98% | 38 | 103 | 188 | 64 | 80 | 164 |
| dRPC | No (measured, not listed here) | 0.670 | 99.96% | 56 | 641 | 1,950 | 153 | 875 | 895 |
Source: Chainstack Labs Research Team, Ethereum RPC provider benchmark, October 2026 (7-day window ending October 8, 2026).
What the benchmark shows:
- Among the three listed providers it measured, Chainstack has the lowest P50 (36 ms), P95 (78 ms) and P99 (117 ms) and the best overall score (0.079). The report puts second-place Quicknode’s score 23% higher.
- Chainstack is not fastest everywhere. In the US, Alchemy (80 ms P95) and Quicknode (98 ms) beat Chainstack (143 ms). In Germany, Quicknode (40 ms) beats Chainstack (47 ms). Chainstack is lowest in Singapore (42 ms), and its best-to-worst region ratio is 3.4x (42 ms in Singapore against 143 ms in the US).
- Chainstack, Quicknode and Alchemy tie on probe success rate at 99.98% (dRPC is 99.96%). This is the share of benchmark probe requests that succeeded over 7 days; it is not platform uptime and not an SLA.
- This article’s practical threshold is a P95 below 500 ms for
eth_calland under 2 seconds for paginatedeth_getLogs. The benchmark’s P95 is an average across methods, so it does not test those thresholds directly, but no regional P95 of the three listed providers measured exceeds 164 ms. RWA teams should compare the region where their compliance polling runs and measureeth_callandeth_getLogsthemselves. - dRPC, which is not one of the 7 listed providers, is degraded in the US (875 ms P95) and Singapore (895 ms P95).
What this benchmark does not show: RouteMesh, Dwellir, Blockdaemon and Infura were not measured, so this page does not claim Chainstack is the fastest of the 7 listed providers. The report covers a single 7-day window, each provider on its first paid plan only (free, higher and dedicated plans can behave differently), and synthetic probes from three cloud regions rather than end-user latency. It excludes eth_subscribe, so it says nothing about WebSocket subscription latency or stability, and this article cites no per-method latency ranges. It does not measure throughput under load, cost or correctness, and because Chainstack chose the scoring formula, a different formula could reorder close pairs.
Before committing to a provider, benchmark from your actual deployment region rather than relying on aggregated averages. p95 and p99 latency are more relevant than median for compliance-critical workflows where a single slow response blocks a transfer.
⚡ Benchmark before you commit: Run a curl timing loop or a short k6 script against each provider’s endpoint from your target region before finalizing your infrastructure choice — even a 100-request sample surfaces meaningful p95 differences for
eth_callandeth_getLogs.
Getting started with RWA on Chainstack
Deploy a production Ethereum endpoint for RWA infrastructure in a few steps and build better with Ethereum on Chainstack:
- Log in to the Chainstack console (or create an account).
- Create a new project
- Select Ethereum as your blockchain protocol
- Choose network: Ethereum Mainnet or Sepolia / Hoodi testnet
- Deploy the node
- Open Access and credentials and copy your HTTPS and WebSocket endpoints
For continuous background indexing workloads — the standard pattern for RWA event monitoring — evaluate Dedicated Nodes or the Unlimited Node add-on after the initial deploy to eliminate per-request billing entirely; both are available inside the same project.
Here is a minimal compliance eligibility check using eth_call via web3.py, the pattern that runs on every investor transfer attempt in an ERC-3643 protocol:
from web3 import Web3
w3 = Web3(Web3.HTTPProvider("YOUR_CHAINSTACK_ENDPOINT"))
# Minimal ABI for ERC-3643 identity registry isVerified()
IDENTITY_REGISTRY_ABI = [
{
"inputs": [{"type": "address", "name": "userAddress"}],
"name": "isVerified",
"outputs": [{"type": "bool"}],
"stateMutability": "view",
"type": "function",
}
]
IDENTITY_REGISTRY_ADDRESS = w3.to_checksum_address("YOUR_IDENTITY_REGISTRY")
registry = w3.eth.contract(address=IDENTITY_REGISTRY_ADDRESS, abi=IDENTITY_REGISTRY_ABI)
def check_investor_eligibility(investor_address: str) -> bool:
return registry.functions.isVerified(
w3.to_checksum_address(investor_address)
).call()
print(check_investor_eligibility("YOUR_WALLET_ADDRESS"))
📖 For the full integration guide, see the Chainstack Ethereum tooling documentation.
Need testnet ETH? Grab some from the Chainstack Sepolia faucet.
🤖 You can also access Chainstack Ethereum RPC directly from Claude, Cursor, Codex, Windsurf, Gemini CLI, GitHub Copilot, Antigravity, Claude.ai, or ChatGPT using Chainstack MCP. For a fuller agent stack — MCP, the Chainstack skill, llms.txt for context ingestion, and WebMCP for agentic browsers — see the Chainstack Agents page.
Conclusion
In our assessment, provider selection for Ethereum RWA in 2026 is primarily a compliance question, not a performance question. Archive access defaults, published certification depth, pricing predictability for eth_call-heavy compliance workflows, and dedicated infrastructure isolation separate the providers here more than raw latency does. The October 2026 latency benchmark above covers only three of the seven providers, so it does not determine the order below.
- For compliance-critical RWA with institutional due diligence requirements: in our assessment, Chainstack (published by Chainstack, see how we compared) — dual SOC 2 Type II + ISO 27001, method-agnostic pricing, archive on paid plans, MEV protection by default
- For teams that want managed failover across many upstream Ethereum providers: RouteMesh — automatic rerouting reduces single-endpoint downtime risk during compliance polling
- For teams that want flat per-response billing and a dedicated archive cluster: Dwellir — trace/debug on paid plans and a 99.99% uptime SLA claimed on its marketing pages (contract terms not verified), but no public SOC 2 or ISO 27001
- For institutional issuers already evaluating custody or staking vendors: Blockdaemon — bundles compliance-grade nodes with custody and tokenization services
- For established Ethereum-native deployments with ConsenSys ecosystem ties: Infura — deep Ethereum protocol familiarity, archive data on every plan
- For developer-centric RWA dApps at early stage: Alchemy — generous free tier and Ethereum tooling depth; CU weighting becomes a constraint at scale
- For teams that specifically need ISO 27001 alongside SOC 2 from a direct competitor: Quicknode — SOC 2 Type II and ISO 27001 published, at a higher list price per unit of throughput
Frequently asked questions
Q: What RPC methods do ERC-3643 RWA protocols call most frequently on Ethereum?
The most frequent methods are eth_call for eligibility checks against the ONCHAINID identity registry on every transfer simulation, eth_getLogs for indexing Transfer and ComplianceCheck events across both the token contract and the identity registry, and eth_subscribe("logs") for real-time event streaming. Archive methods — eth_getStorageAt and debug_traceTransaction — run less frequently but are essential for audit trail reconstruction.
Q: Which Ethereum RPC provider is best for prototyping an RWA compliance workflow before committing to a paid plan?
Chainstack and Alchemy both offer permanent free tiers with no time limit — Chainstack at 3M RU/month with no card required, Alchemy at 30M CUs/month. Quicknode’s entry offering is a one-month trial that converts to a paid plan afterward, not a standalone free tier. RouteMesh and Dwellir both offer lower-friction access for early testing (Dwellir has a free plan with 100,000 responses per day), though neither publishes the compliance certification a production RWA deployment will eventually need.
Q: How do I migrate an existing RWA protocol from Infura or Alchemy to a method-agnostic provider without downtime?
Deploy the new endpoint alongside the existing one, point a canary percentage of traffic at it, and compare eth_call and eth_getLogs responses against your current provider before cutting over fully. Because Chainstack, Infura, and Alchemy all expose standard JSON-RPC, the application code doesn’t need to change — only the endpoint URL and any provider-specific rate-limit handling. Keep the old endpoint live until the new one has run a full settlement cycle without a compliance event gap.
Q: What RPC latency does a production Ethereum RWA protocol need?
RWA infrastructure is consistency-critical rather than ultra-low-latency. A p95 below 500ms for eth_call and under 2 seconds for paginated eth_getLogs across a 5,000-block range is sufficient for most protocols. More important than median latency is the absence of rate-limit interruptions during compliance polling and settlement windows. In Chainstack’s October 2026 Ethereum benchmark (a 7-day, Chainstack-run test of Chainstack, Alchemy, Quicknode and dRPC), the overall P95 was 78 ms for Chainstack, 92 ms for Quicknode and 103 ms for Alchemy, and no regional P95 of those three exceeded 164 ms; these are averages across methods, not eth_call or eth_getLogs figures. Benchmark from your deployment region using the Chainstack performance dashboard before committing.
Q: Which providers have published certifications relevant to RWA vendor due diligence?
Chainstack holds both SOC 2 Type II (December 2025) and ISO 27001 (July 2026). Quicknode and Blockdaemon also publish both SOC 2 Type II and ISO 27001 (Quicknode lists the reports as available on request on its enterprise tier). Alchemy publishes SOC 2 Type II only — not ISO 27001. Infura’s pricing page lists neither; confirm with ConsenSys which reports cover Infura. RouteMesh, as a routing layer, doesn’t publish certifications of its own — the certification that matters is whichever underlying provider ultimately serves the request. We found no SOC 2 or ISO 27001 attestation for Dwellir either, so request documents from it directly.
Q: How much more does method-weighted pricing cost for RWA compliance workloads?
eth_call costs 26 CU on Alchemy versus 1 RU on Chainstack for the same recent-state read. A protocol running 500,000 compliance checks per month against recent state uses 500,000 RUs on Chainstack (roughly $1.25 at the effective plan rate of about $2.45–$2.50 per million) against 13M CUs on Alchemy at 26 CU per call, on this method alone. Total cost impact depends on your full method mix — run your own numbers through the cost comparison rather than assuming a flat multiplier applies to every method. The Alchemy CU figure and the effective plan rate come from Chainstack’s own comparison and are not independently verified on this page.
Additional resources
- How to get an Ethereum RPC endpoint in 2026
- Ethereum tooling documentation — Chainstack Docs
- Ethereum methods — Chainstack Docs
- Enable Debug & Trace APIs for your Ethereum node — Chainstack Docs
- Blockchain infrastructure for RWA
- RPC infrastructure for RWA: EVM node requirements
- More Ethereum tutorials and articles on the Chainstack Blog
- Chainstack pricing
Changelog
We review this comparison regularly so that the information stays accurate, clear, and easy to verify. This page was reviewed to improve how the comparison is structured and sourced.
- October 29, 2026: Added a quick verdict, a Rank and Best for column in the provider comparison table (moved up), a methodology and disclosure section, an Ethereum latency benchmark section (October 2026, covering Chainstack, Alchemy and Quicknode) replacing the previous undated benchmark paragraph, a rate limits and free tiers comparison, alternatives guidance, numbered provider sections with one-line verdicts and official pricing links, a platform uptime statement sourced to the Chainstack status page, replacing the previous uptime statements, corrected vendor pricing, limits and certification details, and wording clarifications.
- May 13, 2026: Original publication.