Solana RPC provider scores for bridges and swap (November 2026)
Data sources: Compare scorecard and the 7-day dashboard, both read Nov 10, 2026. The scorecard page shows no timestamp.
Abstract
Chainstack has the highest score on this scorecard for Solana bridges and swap: 78 out of 100, 5.2 points ahead of Alchemy at 73, with Quicknode at 68. Best for Cost and Coverage: Chainstack. Best for Limits and RPS and Support: Alchemy. Best for Enterprise: Quicknode. The score rests on 76% of the total weight. Chainstack publishes the scorecard and is one of the three providers compared.
Key findings
- Chainstack scores 78, Alchemy 73 and Quicknode 68, built from 76% of the weight.
- Chainstack has the highest Cost score (60 against 34 and 33) at $4.87 per 1M requests, against $16.64 for Quicknode and $12.74 for Alchemy.
- Chainstack has the highest Coverage score (94 against 87 and 77), a pillar weighted 15 for this use case.
- Chainstack does not lead on Reliability: its availability score is 87, against 100 for Alchemy and Quicknode, because Frankfurt success was 99.945% over the 7 days.
- Chainstack does not lead on Limits and RPS (49 against 62 for Alchemy), and has the lowest Support score (12 against 26 and 23).
Verdict and best for
| Rank | Provider | Score | Best for | Weak spot |
|---|---|---|---|---|
| 1 | Chainstack | 78 | Cost (60) and Coverage (94) | Reliability 87, Limits and RPS 49, Support 12 |
| 2 | Alchemy | 73 | Limits and RPS (62), Support (26) | Cost 33, Enterprise 48 |
| 3 | Quicknode | 68 | Enterprise (93) | Limits and RPS 15, Coverage 77 |
Alchemy and Quicknode score 100 on Reliability, and Chainstack, Alchemy and Quicknode all score 100 on Performance.
Contents
- How the score is built
- Solana RPC provider scores: pillar table
- Solana RPC provider comparison by pillar
- Reliability and Performance from the 7-day dashboard
- Conclusion
- Appendix: weights and rules
- Other RPC providers
How the score is built
The score is the weighted mean of eight pillars, each weighted for the Bridges and swap use case. The scorecard scored six pillars on its own. This report computed Reliability and Performance from the 7-day dashboard and combined all eight into one score per provider.
What the scorecard scored
- Cost, Limits and RPS, Coverage, Enterprise, Support and AI-readiness are scored by the scorecard. They are compiled by hand from public pages, and the scorecard says to confirm prices and limits with each provider.
- Limits and RPS is 60% scored. Behavior under a spike (2x the limit) needs a load test and is not scored.
- Chainstack publishes the scorecard and is one of the providers compared.
What this report adds
- Reliability and Performance showed "Not scored" on the scorecard on Nov 10, 2026. This report computed them from the dashboard's 7-day window, read the same day, with the scorecard's published formulas.
- Reliability rests on availability alone, which is 40% of the pillar. Freshness, correctness and a separate error-rate measure were not available. Performance rests on the mean of three regional P95 latencies, 100% of the pillar.
- The six static pillars alone gave 69 for Chainstack, 58 for Alchemy and 51 for Quicknode, the totals the scorecard showed that day. Those totals exclude live Reliability and Performance, so they differ from the scores in this report.
- The score therefore rests on 76% of the total weight.
What the dashboard measures
- Three probe regions: Frankfurt, US West and Singapore. The first paid plan of each provider (the dashboard names Growth for Quicknode and Alchemy, and no plan for Chainstack). No end-user latency and no correctness measurement.
- Availability is the mean of the three regional request success rates, from the scorecard's measurements endpoint. The dashboard snapshot reports one success rate per provider, which is not used for scoring.
- Helius and dRPC appear on the dashboard but are not scored by the scorecard, so they are left out.
How to read the numbers
- Displayed scores are rounded to whole numbers. A gap under 3 points is a tie by the scorecard's rule; the gaps in this report are 5.2 and 4.5 points, so no pair is tied.
- Highest score marks the best value in each row. For prices, latency and failure rates, the lowest value is best.
Solana RPC provider scores: pillar table
| Pillar (weight) | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| Reliability (30), from the 7-day dashboard | 87 | 100 | 100 |
| Cost (15) | 60 | 33 | 34 |
| Performance (15), from the 7-day dashboard | 100 | 100 | 100 |
| Limits and RPS (15) | 49 | 62 | 15 |
| Coverage (15) | 94 | 87 | 77 |
| Enterprise (5) | 88 | 48 | 93 |
| Support (3) | 12 | 26 | 23 |
| AI-readiness (2) | 79 | 79 | 79 |
| Score | 78 | 73 | 68 |
Higher is better; bold marks the highest in each row, including ties.
Solana RPC provider comparison by pillar
Cost and pricing per 1M requests (weight 15)
| Metric | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| $ per 1M requests at the benchmark (70% of pillar), lower is better | $4.87 (53) | $12.74 (25) | $16.64 (17) |
| Bill predictability (20%) | 100 | 70 | 100 |
| Archive, trace and history, $ per 1M (10%), lower is better | $10.65 (30) | $20.80 (11) | $16.64 (17) |
Chainstack's benchmark price is on the Growth plan, Alchemy's on pay as you go and Quicknode's on Accelerate.
Limits and RPS (rate limits, weight 15)
| Metric | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| RPS on a ~$50 plan (40%) | 50 RPS, Growth $49 (23) | 408 RPS, PAYG (93) | 50 RPS, Build $49 (23) |
| Flat-fee or unlimited RPS (20%) | 100: Unlimited Node, 25–500 RPS from $149/mo | 0: Enterprise from 1,000 RPS by contract | 0: Enterprise Dedicated only |
| Behavior under spike (40%) | not scored | not scored | not scored |
Alchemy leads the RPS metric. Chainstack leads the flat-fee metric.
Coverage (supported chains, weight 15)
| Metric | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| Listed chain count (30%) | 98: 93 listed protocols | 90: 75 mainnets | 91: 77 chains |
| Dedicated nodes (10%) | 100: self-serve from Pro | 50: sales only | 50: Enterprise only |
| Region selection (10%) | 100 | 50 | 50 |
| MEV protection or private tx (10%) | 50: Warp, $0.15 per tx, not front-running protection | 100: on by default | 0: not disclosed |
| Archive, WebSocket, gRPC, testnets (10% each) | 100 each | 100 each | 100 each |
Chainstack scores 50 on MEV protection because Warp speeds up sendTransaction and is not front-running protection. Alchemy scores 100.
Enterprise (weight 5)
| Metric | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| SOC 2 Type II (25%) | 100 | 100 | 100 |
| ISO 27001 (10%) | 100 | 0 | 100 |
| SLA of 99.9% or more with credits (25%) | 100: 99.9% | 0: percentage not disclosed | 100: 99.99% |
| SSO, RBAC, audit log (15%) | 67 | 100 | 100 |
| Status page, 90 days of history (10%) | 100 | 0 | 100 |
Quicknode leads the pillar at 93. Chainstack's audit log is not documented.
Support (weight 3)
| Metric | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| Response time, ~$50 plan (50%) | 0: under 24 h, 24×5 | 18: 24 h | 18: 24-hour SLA |
| Response time, ~$500 plan (30%) | 0: under 24 h, 24×5 | 18: 24 h | 46: 8-hour SLA |
| Channels and coverage (20%) | 60 | 60 | 0 |
AI-readiness (weight 2)
| Metric | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| MCP, 5 reference tasks (35%) | 80: 4 of 5 | 40: 2 of 5 | 40: 2 of 5 |
| Skills and plugins (10%) | 50 | 100 | 100 |
| Agent self-serve and payments (15%) | 40: no x402 | 100 | 100 |
All three score 100 on docs for LLMs and machine-readable pricing.
Reliability and Performance from the 7-day dashboard
Computed from the dashboard's 7-day window, read Nov 10, 2026.
| Measure | Chainstack | Alchemy | Quicknode |
|---|---|---|---|
| Availability, mean of three regions, higher is better | 99.982% | 100% | 99.993% |
| Reliability score (weight 30, 40% of pillar) | 87 | 100 | 100 |
| P95 latency, Frankfurt, ms, lower is better | 29.4 | 24.6 | 26.0 |
| P95 latency, US West, ms | 54.4 | 38.3 | 53.3 |
| P95 latency, Singapore, ms | 40.7 | 146.7 | 64.1 |
| P95 latency, mean of three regions, ms | 41.5 | 69.8 | 47.8 |
| Performance score (weight 15) | 100 | 100 | 100 |
Performance scores 100 for any mean at or below 80 ms, so all three providers tie. Alchemy has the lowest P95 in Frankfurt and US West, and Chainstack the lowest mean because Alchemy's Singapore P95 is 146.7 ms.
Conclusion
Chainstack has the highest score on this scorecard for Solana bridges and swap: 78, 5.2 points ahead of Alchemy at 73, on 76% of the total weight. Quicknode scores 68.
Per pillar:
- Reliability: Alchemy and Quicknode, 100 each.
- Cost: Chainstack, 60.
- Performance: a three-way tie at 100.
- Limits and RPS: Alchemy, 62.
- Coverage: Chainstack, 94.
- Enterprise: Quicknode, 93.
- Support: Alchemy, 26.
- AI-readiness: a three-way tie at 79.
A reader who weights Reliability or Limits and RPS above Cost and Coverage should read those rows first.
Appendix: weights and rules
Weights for Bridges and swap: Reliability 30, Cost 15, Performance 15, Limits and RPS 15, Coverage 15, Enterprise 5, Support 3, AI-readiness 2.
- Performance is 100 at a mean regional P95 of 80 ms or less, 0 at 1,500 ms or more, and
100 × ln(1500 / m) / ln(1500 / 80)between. - Reliability is 100 at availability of 99.99% or more, 0 at 99% or less, and
50 × (−log10(1 − a) − 2)between. - Score is the sum of weight × scored share × pillar score, divided by the sum of weight × scored share.
Other RPC providers
The scorecard scores only the providers in this report. These Chainstack comparison lists cover others: