Deribit vs Hyperliquid: which API is more reliable?
Over the 60 days of a 90-day window up24 has measured so far, Hyperliquid was more available — 99.71% against 99.98% of checks answered, and Deribit was quicker — a typical p95 of 64ms against 375ms.
Both venues are probed by the same probes on the same schedule — 90-day window, one request every 5 seconds, percentiles over successful responses only and measured from Lauterbourg, FR. Neither exchange reports any of this; up24 measures it.
| MEASUREMENT | ||
|---|---|---|
Uptime · 90d | 99.71% | 99.98% (better) |
REST checks behind it | 3,007,981 | 3,001,316 |
Days measured | 60 of 90 | 60 of 90 |
Typical p50 | 32ms (better) | 242ms |
Typical p95 | 64ms (better) | 375ms |
Worst day | 93.95% | 99.88% (better) |
up24 incidents · 90d | 173 | 37 (better) |
WS delivering · 24h | 99.95% (better) | 99.88% |
WS connected · 24h | 99.99% (better) | 99.98% |
Reconnects · 24h | 20 (better) | 54 |
Sequence gaps · 24h | 0 | not published |
Longest feed silence · 24h | 5 min 1s | 24s (better) |
Own status feed | deribit-system | statuspage |
Incidents they acknowledged | 0 of 173 | 0 of 37 |
Median gap to their own notice | — | — |
DERIBIT
Over the 60 days of a 90-day window up24 has measured so far, Deribit's REST API answered 99.71% of 3,007,981 checks, with a typical p95 of 64ms and a typical p50 of 32ms. Its worst day was 18 AUG 2026, at 93.95%. up24 opened 173 incidents against it in that window — uptime counts REST checks only, and an incident is as often a WebSocket feed that stalled or a socket that dropped.
Over the last 24 hours Deribit's WebSocket feeds were delivering 99.95% of the time on a socket up24 held open 99.99% of it, across 20 reconnects, with 0 sequence gaps and a longest silence of 5 min 1s.
HYPERLIQUID
Over the 60 days of a 90-day window up24 has measured so far, Hyperliquid's REST API answered 99.98% of 3,001,316 checks, with a typical p95 of 375ms and a typical p50 of 242ms. Its worst day was 15 AUG 2026, at 99.88%. up24 opened 37 incidents against it in that window — uptime counts REST checks only, and an incident is as often a WebSocket feed that stalled or a socket that dropped.
Over the last 24 hours Hyperliquid's WebSocket feeds were delivering 99.88% of the time on a socket up24 held open 99.98% of it, across 54 reconnects, (sequence numbers are not checked for this feed), with a longest silence of 24s.
Uptime is ok checks over total checks, REST only. Both percentiles are medians of daily percentiles over the window rather than the window's own — percentiles do not compose, and the daily rollup is the finest grain that outlives raw retention. Green marks the better column on that row and nothing else: a venue with no measurement never wins one.