Real-time latency profile of the Sgraal preflight pipeline.
12ms
p50 latency
23ms
p95 latency
41ms
p99 latency
~3.5s
Full standard pipeline
Time to first byte. The API responds as soon as the scoring engine completes. This is what your agent waits for.
Full 83-module scoring pipeline time. Includes all analytics modules. Returned in the pipeline_ms field of every preflight response.
| Layer | Modules | Typical ms |
|---|---|---|
| Scoring engine | 10 core components | 0.1ms |
| Security detection | 4 detection layers | 0.0ms |
| Analytics modules | 83 total | ~3,400ms |
| Total | — | ~3,500ms |
Set response_profile: "compact" to skip heavy analytics modules. Compact pipeline: ~100ms total.
informational and reversible actions auto-select compact profile. irreversible and destructive get standard (full pipeline).
from sgraal import SgraalClient result = client.preflight(memory_state=[...]) print(result["_trace"]["duration_ms"]) # pipeline time in ms