Live synthetic monitoring

TrustedRouter Status

Check TrustedRouter availability across regions with live gateway, attestation, SDK, billing, provider fallback, latency, incident history, and router-core uptime signals.

Last updated 2026-09-08T22:05:58Z

All Systems Operational

Router-core synthetic checks are passing for attested reachability, authorization, fallback, and settlement.

Operational · Router Core 99.99% Availability target In region gateway overhead p50: 60 ms Global observed gateway p50: 279 ms Latest synthetic sample 2026-09-08T22:04:05.000Z

Router Core

Attested TLS, authorization, route candidates, provider fallback, and settlement/refund durability.

Operational
Target 99.99% 5m uptime >99.99% 24h uptime 99.9789% 5m burn 0.0x

Control Plane

Dashboard, billing, key management, docs, and public status surfaces.

Operational
Target 99.99% 5m uptime >99.99% 24h uptime >99.99% 5m burn 0.0x

Gateway latency anatomy

Fresh connections include DNS, TCP, and TLS setup. Reused connections show the request path after that setup is paid once.

ConnectionEndpointRegion pathTotal p50DNSTCPTLSGatewaySamples
Fresh connection Global endpoint europe-west4 -> us-central1 387 ms 7 ms 117 ms 136 ms <1 ms 1
Reused connection Global endpoint europe-west4 -> us-central1 104 ms paid once paid once paid once <1 ms 1
Fresh connection EU direct europe-west4 -> europe-west4 60 ms 13 ms 9 ms 7 ms <1 ms 1
Reused connection EU direct europe-west4 -> europe-west4 2 ms paid once paid once paid once <1 ms 1
Fresh connection US Central direct europe-west4 -> us-central1 390 ms 6 ms 141 ms 115 ms <1 ms 1
Reused connection US Central direct europe-west4 -> us-central1 104 ms paid once paid once paid once <1 ms 1
Fresh connection US East direct europe-west4 -> us-east4 303 ms 6 ms 89 ms 95 ms <1 ms 1
Reused connection US East direct europe-west4 -> us-east4 89 ms paid once paid once paid once <1 ms 1

Upstream provider health

Continuously measured per provider from the rotation probe and real traffic over the rolling benchmark set of up to 10,000 samples. Ranked by success rate, then p50 TTFT. Unsupported route and probe-configuration rows are shown separately, not counted as provider downtime. This is an informational provider watch — a flaky upstream model does NOT count against router-core uptime above. Full leaderboard.

ProviderSuccessp50 TTFTError rateTop errorConfig excludedSamples
neurometric 100.00% 953 ms 0% 500
fireworks 100.00% 990 ms 0% 266
stepfun 100.00% 1014 ms 0% 32
mistral 100.00% 1066 ms 0% 30
cloudflare-workers-ai 100.00% 1085 ms 0% 24
grok 100.00% 1115 ms 0% 30
scaleway 100.00% 1204 ms 0% 7 probe_config_error 23
digitalocean 100.00% 1239 ms 0% 41
wandb 100.00% 1335 ms 0% 3 probe_config_error 21
engy 100.00% 1370 ms 0% 19
google-vertex 100.00% 1412 ms 0% 28
io-net 100.00% 1461 ms 0% 42
atlas-cloud 100.00% 1495 ms 0% 2 provider_error 4
databricks 100.00% 1497 ms 0% 33
thinkingmachines 100.00% 1505 ms 0% 37
relace 100.00% 1513 ms 0% 43
openai 100.00% 1529 ms 0% 35
azure 100.00% 1535 ms 0% 4 probe_config_error 22
friendli 100.00% 1550 ms 0% 25
telnyx 100.00% 1575 ms 0% 2 probe_config_error 43
google-ai-studio 100.00% 1650 ms 0% 38
nscale 100.00% 1708 ms 0% 23
anthropic 100.00% 1803 ms 0% 39
alibaba 100.00% 1842 ms 0% 37
zero-g 100.00% 2083 ms 0% 47
minimax 100.00% 2434 ms 0% 34
perplexity 100.00% 2437 ms 0% 33
zai 100.00% 2467 ms 0% 31
together 100.00% 2718 ms 0% 7 probe_config_error 99
arcee 100.00% 4211 ms 0% 38
kimi 100.00% 4249 ms 0% 155
sakana 100.00% 6070 ms 0% 15 probe_config_error 19
nebius 100.00% 22669 ms 0% 92
jina 100.00% 0% 5
cerebras 99.51% 1322 ms 0.5% TimeoutError 22 probe_config_error 203
deepseek 97.98% 2214 ms 2.0% provider_error 495
baseten 97.98% 3664 ms 2.0% provider_error 495
deepinfra 97.78% 1397 ms 2.2% TimeoutError 45
siliconflow 97.44% 1883 ms 2.6% TimeoutError 39
parasail 96.97% 1817 ms 3.0% provider_error 33
wafer 96.67% 2581 ms 3.3% TimeoutError 30
gmi 96.55% 3751 ms 3.5% TimeoutError 1 unsupported_route 29
pearl 95.83% 1409 ms 4.2% provider_error 24
lightning 95.45% 1452 ms 4.5% provider_error 22
tinfoil 94.78% 2903 ms 5.2% provider_error 4 probe_config_error 115
venice 94.44% 2196 ms 5.6% TimeoutError 54
xiaomi 92.68% 2670 ms 7.3% TimeoutError 41
featherless 91.30% 2189 ms 8.7% empty_stream 10 probe_config_error 23
novita 87.88% 1712 ms 12.1% TimeoutError 33
inceptron 87.50% 2887 ms 12.5% provider_error 32
makora 84.21% 1460 ms 15.8% TimeoutError 10 probe_config_error 19
akashml 82.76% 2215 ms 17.2% empty_stream 29
morph 60.47% 2519 ms 39.5% provider_error 43
nvidia-nim 50.00% 3785 ms 50.0% TimeoutError 36
chutes 42.31% 11955 ms 57.7% provider_error 26
phala 41.94% 3945 ms 58.1% provider_error 31
near-ai 6.25% 6553 ms 93.8% provider_error 32

Error budget burn

Burn rate means how quickly failures are spending the allowed error budget. Router Core is the paging signal. Upstream reliability is measured separately for every provider above.

Warning Router Core · 6h · 9.62x burn 1 bad / 1040 samples

Components

Each row is backed by live US/EU synthetic checks. Prompt and output content are not stored.

Operational Degraded Outage No data

Canonical API

api.trustedrouter.com attested TLS reachability and trust checks.

Operational
24h uptime 99.9474% Latency groups 2 Last check 2026-09-08T22:03:42.000Z
attestation_nonce · Global endpoint · europe-west4 -> us-central1 p50 1322 ms tls_health · Global endpoint · europe-west4 -> us-central1 p50 279 ms

US Central Regional API

US Central attested TLS reachability and trust checks.

Operational
24h uptime 99.9474% Latency groups 2 Last check 2026-09-08T22:03:43.000Z
attestation_nonce · US Central direct · europe-west4 -> us-central1 p50 1996 ms tls_health · US Central direct · europe-west4 -> us-central1 p50 403 ms

US East Regional API

US East attested TLS reachability and trust checks.

Operational
24h uptime 99.7368% Latency groups 2 Last check 2026-09-08T22:03:42.000Z
attestation_nonce · US East direct · europe-west4 -> us-east4 p50 1158 ms tls_health · US East direct · europe-west4 -> us-east4 p50 363 ms

EU Regional API

EU attested TLS reachability and trust checks.

Operational
24h uptime 99.9474% Latency groups 2 Last check 2026-09-08T22:03:42.000Z
attestation_nonce · EU direct · europe-west4 -> europe-west4 p50 821 ms tls_health · EU direct · europe-west4 -> europe-west4 p50 60 ms

Attestation

Nonce and digest verification for public attested gateways.

Operational
24h uptime 99.7895% Latency groups 4 Last check 2026-09-08T22:03:43.000Z
attestation_nonce · Global endpoint · europe-west4 -> us-central1 p50 1322 ms attestation_nonce · EU direct · europe-west4 -> europe-west4 p50 821 ms attestation_nonce · US Central direct · europe-west4 -> us-central1 p50 1996 ms attestation_nonce · US East direct · europe-west4 -> us-east4 p50 1158 ms

Billing and Settlement

Authorize, settle, and accounting path used by the gateway.

Operational
24h uptime >99.99% Latency groups 2 Last check 2026-09-08T22:03:46.000Z
gateway_authorize · Control plane · europe-west4 p50 3260 ms gateway_settle · Control plane · europe-west4 p50 2717 ms

Provider Fallback

Fail-first route selection and rollover to the next healthy provider.

Operational
24h uptime >99.99% Latency groups 1 Last check 2026-09-08T22:04:05.000Z
provider_fallback · Control plane · europe-west4 p50 5429 ms

Image Generation

Public attested Gemini image generation and binary image validation.

Operational
24h uptime >99.99% Latency groups 0 Last check 2026-09-08T18:19:58.000Z

Model Inference

End-to-end model calls through the attested gateway: OpenAI SDK and Responses round-trips with verified output.

Operational
24h uptime 99.9868% Latency groups 8 Last check 2026-09-08T22:04:00.000Z
openai_sdk_pong · Global endpoint · europe-west4 -> us-central1 p50 1454 ms responses_pong · Global endpoint · europe-west4 -> us-central1 p50 1192 ms openai_sdk_pong · EU direct · europe-west4 -> europe-west4 p50 4785 ms responses_pong · EU direct · europe-west4 -> europe-west4 p50 4695 ms openai_sdk_pong · US Central direct · europe-west4 -> us-central1 p50 3532 ms responses_pong · US Central direct · europe-west4 -> us-central1 p50 2898 ms openai_sdk_pong · US East direct · europe-west4 -> us-east4 p50 2506 ms responses_pong · US East direct · europe-west4 -> us-east4 p50 2378 ms

Router Core uptime

Precomputed Router Core rollups. Control Plane and upstream providers are measured separately.

WindowStatusSamplesGroups
5m up 6 5
24h up 4740 10
48h down 9530 10

Recent Events

Samples that were not operational in the last 24 hours.

Major outage
US Central Regional API

2026-09-08T20:00:00Z · attestation_nonce · us-central1 · hourly rollup · 1 failed / 21 · attestation_http_500

Major outage
Canonical API

2026-09-08T19:00:00Z · attestation_nonce · europe-west4 · hourly rollup · 1 failed / 20 · attestation_http_500

Major outage
US East Regional API

2026-09-08T18:00:00Z · attestation_nonce · europe-west4 · hourly rollup · 1 failed / 20 · attestation_http_500

Major outage
US East Regional API

2026-09-08T18:00:00Z · attestation_nonce · us-central1 · hourly rollup · 1 failed / 20 · attestation_http_500

Major outage
US East Regional API

2026-09-08T16:00:00Z · attestation_nonce · us-central1 · hourly rollup · 2 failed / 21 · attestation_http_500

Major outage
US East Regional API

2026-09-08T16:00:00Z · attestation_nonce · europe-west4 · hourly rollup · 1 failed / 21 · attestation_http_500

Major outage
EU Regional API

2026-09-08T16:00:00Z · attestation_nonce · us-central1 · hourly rollup · 1 failed / 21 · attestation_http_500

Major outage
Model Inference

2026-09-08T07:00:00Z · responses_pong · us-central1 · hourly rollup · 1 failed / 22 · pong_mismatch

Every cloud runs its own probes and builds this page from its own samples, so these are independent reports rather than copies of one. If the cloud serving this page is the thing that is down, the others keep answering. Every cloud also watches every other cloud's page each pass and flags one that stops reporting.

View fleet health Google Cloud · this page AWS status Azure status
Workspace access

Sign in

Choose a sign in method to access your TrustedRouter workspace.

By signing in you agree to the terms of service and privacy policy.