[Observability] Grafana dashboard for TTFT + ITL #38

Open
opened 2026-09-11 01:27:28 +00:00 by poimen · 0 comments
Member

No visibility into LLM inference latency. Create Grafana dashboard tracking: Time-to-First-Token (TTFT), Inter-Token Latency (ITL), KV cache utilization %, request queue depth, active sequences. Files: k8s/infra/monitoring/, k8s/infra/logging/grafana-values.yaml

No visibility into LLM inference latency. Create Grafana dashboard tracking: Time-to-First-Token (TTFT), Inter-Token Latency (ITL), KV cache utilization %, request queue depth, active sequences. Files: k8s/infra/monitoring/, k8s/infra/logging/grafana-values.yaml
poimen added this to the LLM Production Hardening milestone 2026-09-11 01:27:28 +00:00
Sign in to join this conversation.