Hyperliquid RPC provider scores for production DApps (November 2026)
Data sources: the Chainstack Compare scorecard (Hyperliquid, Production dApp preset) and the 7-day Compare dashboard, both read on Nov 10, 2026. The scorecard shows no timestamp.
Abstract
On the Production dApp weights, Chainstack scores 81 out of 100 for Hyperliquid, Quicknode 68 and Alchemy 66, on 81% of the total weight. Chainstack is best for cost and limits, Alchemy for reliability and support, and Quicknode for enterprise. Reliability and Performance come from the 7-day dashboard.
Key findings
- Chainstack scores 81, 12.5 points ahead of Quicknode at 68. Quicknode and Alchemy (66) are 2.6 points apart, a tie by the page's rule of under 3 points.
- Cost is the widest gap: Chainstack costs $2.62 per 1M requests on the page's EVM dApp mix, against $12.37 for Quicknode and $14.46 for Alchemy.
- Alchemy leads Reliability with 100, from 100% availability. Chainstack scores 82 (99.977%) and Quicknode 77 (99.972%).
- Performance does not separate the providers: mean P95 latency is 29 ms for Chainstack, 37 ms for Alchemy and 73 ms for Quicknode, and all three score 100.
- Quicknode leads Enterprise (93 against 88 for Chainstack) and Alchemy leads Support (26 against 12 for Chainstack).
Verdict and best for
| Rank | Provider | Score (out of 100) | Gap to next |
|---|---|---|---|
| 1 | Chainstack | 81 | 12.5 points |
| 2 | Quicknode | 68 | 2.6 points (tie with Alchemy) |
| 3 | Alchemy | 66 | none |
| Best for | Provider | Score |
|---|---|---|
| Highest score on this scorecard | Chainstack | 81 |
| Cost | Chainstack | 73 |
| Limits and RPS | Chainstack | 85 |
| Coverage | Chainstack | 82 |
| Reliability | Alchemy | 100 |
| Enterprise | Quicknode | 93 |
| Support | Alchemy | 26 |
| Performance | Chainstack, Quicknode, Alchemy (tie) | 100 |
| AI-readiness | Chainstack, Quicknode, Alchemy (tie) | 79 |
Contents
- How the score is built
- Hyperliquid RPC provider scores: pillar table
- Provider summaries
- Hyperliquid RPC provider comparison by pillar
- Reliability and Performance from the 7-day dashboard
- Conclusion
- Appendix: weights and rules
- Related reports and other RPC providers
How the score is built
The score is the page's weighted mean over eight pillars, with each pillar's weight scaled by the share of it that has data. Higher is better for every score.
What the scorecard scored
- Chainstack publishes the scorecard and is one of the three providers compared.
- Cost, Limits and RPS, Coverage, Enterprise, Support and AI-readiness are compiled by hand from public pages. The page says to confirm prices and limits with each provider.
- Limits and RPS is 60% scored. Behavior under a spike (a 2x load test) is not scored.
- Rounded scores can differ by 1 point from an exact recomputation. A gap under 3 unrounded points is a tie by the page's rule.
What this report adds
- Reliability and Performance were not scored by the page when read, so both are computed from the 7-day dashboard window read on Nov 10, 2026. The page's own total over its six static pillars was Chainstack 73, Quicknode 54 and Alchemy 45. That total excludes live Reliability and Performance, which is why it differs from the score in this report.
- Reliability rests on availability alone, which is 40% of the pillar. Freshness and correctness are not scored, so Reliability counts for 10 of its 25 weight points.
- Performance rests on mean regional P95 latency, 100% of the pillar.
- In total, 81% of the weight is scored.
What the dashboard measures
- Three probe regions (Frankfurt, US West and Tokyo), the first paid plan of each provider, and the 7-day window. The dashboard has no end-user latency or correctness measurement.
- Availability is request success, not time-based uptime. The score uses the mean of the three regional success rates from the page's measurements endpoint.
- The dashboard also lists dRPC. The scorecard does not score it, so this report leaves it out.
Hyperliquid RPC provider scores: pillar table
Chainstack scores 81, Quicknode 68 and Alchemy 66. Bold marks the highest score in each row. A row marked "7-day dashboard" was computed for this report.
| Pillar (weight) | Chainstack | Quicknode | Alchemy | Leader |
|---|---|---|---|---|
| Reliability (25), 7-day dashboard, availability only | 82 | 77 | 100 | Alchemy |
| Cost (20) | 73 | 41 | 30 | Chainstack |
| Performance (20), 7-day dashboard | 100 | 100 | 100 | Tie |
| Limits and RPS (10), 60% scored | 85 | 15 | 59 | Chainstack |
| Coverage (5) | 82 | 80 | 71 | Chainstack |
| Enterprise (10) | 88 | 93 | 48 | Quicknode |
| Support (5) | 12 | 23 | 26 | Alchemy |
| AI-readiness (5) | 79 | 79 | 79 | Tie |
| Score | 81 | 68 | 66 | Chainstack |
Provider summaries
1. Chainstack, 81
Verdict: highest score on this scorecard, led by Cost (73), Limits and RPS (85) and Coverage (82). Trade-offs: Support scores 12, the lowest pillar, and gRPC and streams on Hyperliquid score 0.
2. Quicknode, 68
Verdict: best for Enterprise (93) on this scorecard. Trade-offs: Limits and RPS scores 15, with 50 RPS on its ~$50 plan, and Cost scores 41 at $12.37 per 1M requests.
3. Alchemy, 66
Verdict: best for Reliability (100) and Support (26). Trade-offs: Enterprise scores 48 and Cost 30 at $14.46 per 1M requests. Alchemy and Quicknode are a tie by the page's rule.
Hyperliquid RPC provider comparison by pillar
Higher is better for scores. For prices, a lower value is better. Bold marks the best value in each row.
Cost (weight 20)
| Metric (share of pillar) | Chainstack | Quicknode | Alchemy |
|---|---|---|---|
| $ per 1M requests, EVM dApp mix, compute only (70%) | 72 ($2.62, Growth) | 26 ($12.37, Build) | 21 ($14.46, Pay As You Go) |
| Bill predictability (20%) | 100 | 100 | 70 |
| Archive, trace and history, $ per 1M, EVM indexer (10%) | 30 ($10.65) | 24 ($13.32) | 12 ($19.97) |
| Pillar score | 73 | 41 | 30 |
The benchmark is 18.69M calls per month. Alchemy loses predictability points because its compute units run from 10 to 60 CU depending on the method.
Limits and RPS (weight 10, 60% scored)
| Metric (share of pillar) | Chainstack | Quicknode | Alchemy |
|---|---|---|---|
| RPS on a ~$50 plan (40%) | 77 (250 RPS, Growth $49) | 23 (50 RPS, Build $49) | 89 (360 RPS, Pay As You Go) |
| Behavior under spike (40%) | not scored | not scored | not scored |
| Flat-fee or unlimited RPS (20%) | 100 (Unlimited Node, 25–500 RPS, from $149/mo) | 0 | 0 |
| Pillar score | 85 | 15 | 59 |
Alchemy leads the RPS row. Chainstack leads the pillar because of the flat-fee row.
Coverage (weight 5)
| Metric (share of pillar) | Chainstack | Quicknode | Alchemy |
|---|---|---|---|
| Listed chain count (30%) | 98 (93 listed protocols) | 91 (77 chains) | 90 (75 mainnets) |
| Archive on Hyperliquid (8.8%) | 100 (Growth and above) | 100 (all plans) | 100 (all plans) |
| debug and trace (8.8%) | 100 | 100 | 50 |
| WebSocket (8.8%) | 100 | 100 | 100 |
| gRPC and streams (8.8%) | 0 (HyperEVM WebSocket only) | 100 | 50 |
| Dedicated nodes (8.8%) | 100 | 50 | 50 |
| Region selection (8.8%) | 100 | 50 | 50 |
| MEV protection and private transactions (8.8%) | 0 | 0 | 0 |
| Testnets and faucet (8.8%) | 100 | 100 | 100 |
| Pillar score | 82 | 80 | 71 |
Quicknode leads the gRPC and streams row. Chainstack scores 0 there and for MEV protection, where all three providers score 0.
Enterprise (weight 10)
| Metric (share of pillar) | Chainstack | Quicknode | Alchemy |
|---|---|---|---|
| SOC 2 Type II (25%) | 100 | 100 | 100 |
| ISO 27001 (10%) | 100 | 100 | 0 |
| SLA of 99.9% or higher with credits (25%) | 100 | 100 | 0 |
| SSO, RBAC and audit log (15%) | 67 (audit log not documented) | 100 (Enterprise only) | 100 (Enterprise only) |
| Single-tenant and jurisdiction (15%) | 50 | 50 | 50 |
| Status page with 90 days of history (10%) | 100 | 100 | 0 |
| Pillar score | 88 | 93 | 48 |
Support (weight 5)
| Metric (share of pillar) | Chainstack | Quicknode | Alchemy |
|---|---|---|---|
| Response time, ~$50 plan (50%) | 0 (under 24 h, 24×5, up to 72 h) | 18 (24 h) | 18 (24 h) |
| Response time, ~$500 plan (30%) | 0 (under 24 h, 24×5, up to 72 h) | 46 (8 h) | 18 (24 h) |
| Channels and coverage (20%) | 60 | 0 | 60 |
| Pillar score | 12 | 23 | 26 |
Support is Chainstack's lowest pillar on this scorecard.
AI-readiness (weight 5)
| Metric (share of pillar) | Chainstack | Quicknode | Alchemy |
|---|---|---|---|
| MCP, 5 reference tasks (35%) | 80 (4 of 5) | 40 (2 of 5) | 40 (2 of 5) |
| Docs for LLMs (25%) | 100 | 100 | 100 |
| Machine-readable pricing (15%) | 100 | 100 | 100 |
| Skills and plugins, within 90 days (10%) | 50 | 100 | 100 |
| Agent self-serve and payments (15%) | 40 | 100 | 100 |
| Pillar score | 79 | 79 | 79 |
Reliability and Performance from the 7-day dashboard
The window is 7 days, read on Nov 10, 2026, with no flags set. Regions are Frankfurt, US West and Tokyo.
Performance
Mean P95 latency per provider, lower is better. The score is 100 at or under 80 ms.
| Provider | Frankfurt (ms) | US West (ms) | Tokyo (ms) | Mean (ms) | Score |
|---|---|---|---|---|---|
| Chainstack | 19.2 | 46.7 | 21.1 | 29.0 | 100 |
| Alchemy | 32.1 | 25.5 | 52.7 | 36.8 | 100 |
| Quicknode | 19.7 | 78.2 | 122.4 | 73.4 | 100 |
All three means fall under the 80 ms ceiling, so Performance does not change the ranking. Quicknode's Tokyo P95 of 122.4 ms is above 80 ms, and the mean hides that.
Reliability
Availability is the mean of three regional success rates, higher is better.
| Provider | Frankfurt | US West | Tokyo | Mean | Score |
|---|---|---|---|---|---|
| Alchemy | 100% | 100% | 100% | 100% | 100 |
| Chainstack | 99.932% | 100% | 100% | 99.977% | 82 |
| Quicknode | 100% | 100% | 99.915% | 99.972% | 77 |
Chainstack's lower score comes from 99.932% success in Frankfurt. Quicknode's comes from 99.915% in Tokyo.
Conclusion
Chainstack has the highest score on this scorecard for Hyperliquid and the Production dApp preset: 81, 12.5 points ahead of Quicknode at 68, on 81% of the total weight. Quicknode and Alchemy (66) are tied for second.
- Cost: Chainstack, 73.
- Limits and RPS: Chainstack, 85.
- Coverage: Chainstack, 82.
- Reliability: Alchemy, 100.
- Enterprise: Quicknode, 93.
- Support: Alchemy, 26.
- Performance and AI-readiness: three-way ties at 100 and 79.
A team that weights Reliability or Enterprise more heavily than the Production dApp preset does should read those rows first.
Appendix: weights and rules
| Pillar | Weight | Share scored |
|---|---|---|
| Reliability | 25 | 40% |
| Cost | 20 | 100% |
| Performance | 20 | 100% |
| Limits and RPS | 10 | 60% |
| Coverage | 5 | 100% |
| Enterprise | 10 | 100% |
| Support | 5 | 100% |
| AI-readiness | 5 | 100% |
- Score = sum of weight x scored share x pillar score, divided by the sum of weight x scored share.
- Performance: 100 at a mean P95 of 80 ms or less, 0 at 1,500 ms or more, and 100 x ln(1500 / m) / ln(1500 / 80) between.
- Reliability: 100 at availability of 0.9999 or more, 0 at 0.99 or less, and 50 x (−log10(1 − a) − 2) between.
Related reports and other RPC providers
- Hyperliquid RPC Provider Benchmark: Latency and Uptime (October 2026): a Chainstack Labs regional latency and reliability benchmark. Its method and window differ from this scorecard.
The scorecard scores only the providers in this report. These Chainstack lists cover others.