All providers · Provider record

As of Jun 9, 21:58 UTC · revalidates every 60 seconds · pipeline ingests every 5 minutes.

Impact-weighted uptime

Replicate published 2.5 incidents per 30 days on average to its status feed across the last 6 months (n=15). We surface this number instead of classifying providers because uptime feeds vary in publishing volume for reasons we cannot judge from outside; compare the uptime numbers below across providers with this denominator in mind.

7 days
99.50%
n=1 live
30 days
98.49%
n=5 (2 live, 3 historical)
90 days
99.28%
n=8 (2 live, 6 historical)

Uptime weighted by per-incident UIS over the window, computed from incidents this provider published to its own feed. n splits into live (observed in real time) and historical (re-derived from the provider's own archive). Not directly comparable across providers — see the feed-volume disclosure above. How this is computed.

Multi-region reachability + latency

Multi-region probe · last 30 min
3/3 reachable
iad · Washington DC
416 ms
HTTP 200 · 14 min ago
sfo · San Francisco
589 ms
HTTP 200 · 6 min ago
fra · Frankfurt
428 ms
HTTP 200 · 4 min ago

Probes hit the provider's status page from each region every 15 minutes. Per-region latency variance is a signal worth watching even when the provider's own status feed reports operational. When ≥1 region reports unreachable while ≥1 other reports reachable, the disagreement also feeds the §6.1 multi-source confirmation gate.

Published SLA compliance

No published SLA

Replicate does not publish a public Service Level Agreement for its API surface. SLA-compliance verdicts are computed only for providers that publish their target uptime, so this provider has none.

Component classification · 4 rules

primary inference3 rules
auxiliary1 rule

Components classified by class (primary inference, secondary API, auxiliary). Flagship models tracked: 0.

Latency probe · parallel-run, methodology v1.0

No latency probe data captured yet for this provider. The probe runs every 15 minutes once an API key is configured. Methodology v1.0 §4.10 (parallel-run, not used in UIS).

Alt-signal observations · last 24 hours

Unconfirmed signal observations from the last 1440 minutes. None are published incidents; they corroborate confirmed incidents and feed the multi-source gate. Single-source signal is intentionally not auto-posted.

Incident history

UISTitleComponentsStarted (UTC)DurationStatusConf.
76We're seeing long setup times and high contention for models on some L40S and H200 clusters.2026-06-03 17:591h 7mresolvedconfirmed
39Degraded performance on flux-2-klein-4b2026-05-28 12:301h 47mresolvedunconfirmed
40Prediction and Training status updates delayedhistorical record2026-05-21 21:381h 50mresolvedhistorical
99Constrained H100 capacityhistorical record2026-05-21 15:096h 56mresolvedhistorical
41Constrained capacity for H100 hardwarehistorical record2026-05-12 15:274h 16mresolvedhistorical
24Degraded A100 hardwarehistorical record2026-04-19 14:3626mresolvedhistorical
67A100 capacity unavailable during storage maintenancehistorical record2026-04-09 16:4347mresolvedhistorical
41Downstream errors for Black Forest Labs modelshistorical record2026-03-23 15:579h 37mresolvedhistorical
17Degraded performance on Flux Schnellhistorical record2026-03-10 12:565h 21mresolvedhistorical
73Model Predictions Stuck at "Starting"historical record2026-02-20 12:131h 0mresolvedhistorical
17Increased setup failures for T4 modelshistorical record2026-01-26 07:337h 44mresolvedhistorical
36Predictions and training unavailable for multiple modelshistorical record2026-01-20 22:531h 30mresolvedhistorical
41Flux Schnell unavailablehistorical record2026-01-20 20:583h 9mresolvedhistorical
99Prediction Errorshistorical record2026-01-15 07:382h 56mresolvedhistorical
41High demand for H100 hardware typehistorical record2025-12-18 19:0634h 4mresolvedhistorical
32Limited availability of L40S hardwarehistorical record2025-12-11 20:381h 11mresolvedhistorical
100Global network outagehistorical record2025-11-18 11:576h 45mresolvedhistorical
17sora-2-pro currently unavailablehistorical record2025-11-13 03:12106h 52mresolvedhistorical
99Downstream Service Disruptionhistorical record2025-10-29 19:0421h 1mresolvedhistorical
99Luma models not runninghistorical record2025-10-22 14:051h 59mresolvedhistorical
41Intermittent issues with `cog push` with large imageshistorical record2025-10-21 11:072h 11mresolvedhistorical
100Replicate Platform Outagehistorical record2025-10-20 19:287h 18mresolvedhistorical
41Widespread service degradationhistorical record2025-10-20 14:3712h 7mresolvedhistorical
99Heygen models outagehistorical record2025-09-30 17:3747h 18mresolvedhistorical
99Google Models are downhistorical record2025-09-29 18:262h 22mresolvedhistorical
40Low inbound network transfer speed affecting multiple systemshistorical record2025-09-29 02:481h 55mresolvedhistorical
10Users unable to purchase prepaid credithistorical record2025-09-26 22:4433mresolvedhistorical