latenzy · model latency comparison
Fair 750 downloads0 views0 commentsRevision 1Created Converted from Grafana dashboard #25642 by dreamydate1524
How to import
In Datadog: Dashboards → New Dashboard → ⚙ Configure → Import dashboard JSON → paste this JSON.
Reactions
Per-model LLM latency (TTFT, total, throughput, rate-limit pressure) for Claude, OpenAI, and Gemini models, measured by the latenzy synthetic prober. Compare model A vs model B, and the same model over different endpoints (direct / Bedrock / Vertex).
Conversion quality
Fair 75Share of panels whose queries were translated without loss. Partial and unsupported panels keep the original PromQL in a note widget.
- Native0
- OpenMetrics4
- Partial1
- Unsupported1
Metrics without a native Datadog mapping
These metric names were kept in OpenMetrics naming. Adjust them if you collect with a native Datadog integration.
latenzy:ttft_seconds:p95_1h ×2latenzy:request_duration_seconds:p95_1hlatenzy:tokens_per_second:p50_1hlatenzy:probe_failure_ratio:1hlatenzy:rate_limited_ratio:1hUnsupported panels
- Probe staleness · seconds since last success
timeseries
Revisions
- Revision 1Converted from grafana.com revision 1Download JSON
Rate this dashboard
Sign in to rate.0 comments
No comments yet.