API Performance Dashboard
Latency, uptime, and throughput metrics for the Sentiqra API. Figures are an internal measurement snapshot, refreshed periodically - not a live feed. Live availability is on the status page.
Snapshot: 2026-09-18 14:00 UTC
Note: Latency, uptime, and per-product figures are an internal measurement snapshot taken over the window stated below - not a live feed. Peak RPS is a load-test result, not production traffic. Production performance may vary with network conditions, payload size, and concurrent load. Published targets apply to every plan and are not contractual commitments. Contractual SLAs with service-credit remedies are available under Enterprise agreements only. Live availability is on the status page. Live status page →
How this was measured
Published so the figures below can be checked against the approved telemetry aggregates.
| Measurement period | 2026-09-18 00:00-14:00 UTC (14 hours) |
| Region | us-east-1 (sole production region at snapshot time) |
| Request characteristics | Production /v1/analyze traffic: single-analyzer and multi-analyzer requests, text up to 4,096 Unicode code points |
| Analyzers measured | All 12 production analyzers were serving during the window; per-product rows list the four highest-volume services |
| Sample size | ~60.5M analyzed requests during the window |
| Error treatment | Failed and timed-out requests count against uptime windows; latency percentiles are computed over completed responses only |
| Latency scope | Server-side: measured from request receipt to response emission inside the API. Excludes client network time, TLS setup, and edge propagation |
| Data source | Figures are computed from approved request-telemetry aggregates and fixed at snapshot time; the same aggregates reproduce every number on this page |
Latency (us-east-1)
Server-side percentiles over completed responses in the measurement window - not live.
detection
< 500ms p95
languages
< 200ms p95
taxonomy
< 300ms p95
health
< 50ms p95
Uptime
Trailing windows ending at the snapshot timestamp - "24h" is the snapshot day.
24h
100%
7d
99.97%
30d
99.95%
90d
99.93%
Throughput
Typical RPS is the snapshot-window average; peak RPS is a load-test result, not production traffic. Fleet utilization and total capacity ceilings are internal sizing data and are not published.
Typical RPS (snapshot)
~1,200
Peak RPS (load test)
8,500 (load test)
By Region
Measured at snapshot time; planned regions have no measurements.
| Region | p50 | p95 | Status |
|---|---|---|---|
| us-east-1 | 180ms | 412ms | Healthy |
| us-west-2 (planned) | - | - | Planned |
| eu-west-1 (planned) | - | - | Planned |
By Product
Measured at snapshot time; request counts are totals inside the measurement window above, not today.
| Product | p50 | p95 | Requests in window |
|---|---|---|---|
| Trust & Safety | 180ms | 412ms | 36.0M |
| Cybersecurity | 210ms | 480ms | 14.0M |
| Crisis Intelligence | 195ms | 440ms | 7.0M |
| Market Intelligence | 220ms | 490ms | 3.5M |
| Surface | Target | Scope |
|---|---|---|
Analysis (/v1/analyze) |
p95 < 500ms | Every plan |
| Languages | p95 < 200ms | Every plan |
| Taxonomy | p95 < 300ms | Every plan |
| Health | p95 < 50ms | Every plan |
| Availability | 99.9% monthly | Every plan |
Targets describe what we engineer and measure against. Missing a target does not by itself create a contractual remedy.
Contractual SLAs with defined measurement windows, exclusions, and service-credit remedies are available under Enterprise agreements. Remedy terms, scope, and credit tiers are stated in the agreement itself - not on this page. Support response commitments are in the procurement index under Support Terms.
To map the layers: the status page shows live availability, this page shows observed history, targets show intent, and the Enterprise agreement is the only place remedies exist.
Benchmark Results
Per-language Top-3 Hit Rate
Benchmark Methodology
How we measure Top-3 Hit Rate
Playground
Test detection live - no API key
API Reference
Full endpoint documentation
Pricing
From Free to Enterprise
Trust Center
Security & compliance
Quickstart Guide
First API call in under 5 minutes
Request access5,000 analyzer runs/month · No card required