mirror of https://github.com/BerriAI/litellm.git synced 2026-10-05 02:41:56 +00:00

History

yucheng 48b25e448d fix(grafana): sum redis circuit breaker state across workers instead of taking the max Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>		2026-09-17 09:44:26 +00:00
..
dashboard_all_metrics	fix(grafana): sum redis circuit breaker state across workers instead of taking the max	2026-09-17 09:44:26 +00:00
dashboard_genai_otel	feat(cookbook): add a Grafana dashboard for the OTel GenAI metrics (#35159 )	2026-07-30 17:29:40 +00:00
dashboard_v2	feat(grafana): add all-metrics dashboard and fix stale dashboard_v2 gauges	2026-09-17 07:50:36 +00:00
readme.md	feat(grafana): add all-metrics dashboard and fix stale dashboard_v2 gauges	2026-09-17 07:50:36 +00:00

readme.md

Contains LiteLLM maintained grafana dashboard

This folder contains the json for creating Grafana Dashboards

LiteLLM GenAI Dashboard (OpenTelemetry)

Charts the gen_ai.* metrics from the OpenTelemetry v2 integration: spend, tokens, request rate, and latency percentiles by model. Separate from the dashboards below, which chart the litellm_* Prometheus metrics.

LiteLLM All Prometheus Metrics dashboard

Every litellm_* Prometheus metric family the proxy can emit (134 families, 95 panels) grouped by theme: traffic, latency, spend and tokens, cache, deployments, rate limits, budgets, guardrails, MCP, managed files and batches, users and teams, plus the Redis circuit breaker, spend log cleanup and prometheus_system service metrics. Start here if you want everything on one screen; see its readme for import steps and which panels need a feature enabled before they show data

LiteLLM v2 Dashboard

A compact view of proxy request rate, failures, latency and the top remaining-request / remaining-token gauges per model group

Pre-Requisites

Setup LiteLLM Proxy Prometheus Metrics https://docs.litellm.ai/docs/proxy/prometheus