Skip to main content

Benchmarks

Benchmarks for LiteLLM Gateway (Proxy Server)

Locust Settings:

  • 2500 Users
  • 100 user Ramp Up

Basic Benchmarks

Overhead when using a Deployed Proxy vs Direct to LLM

  • Latency overhead added by LiteLLM Proxy: 107ms
MetricDirect to Fake EndpointBasic Litellm Proxy
RPS11961133.2
Median Latency (ms)33140

Logging Callbacks

GCS Bucket Logging

Using GCS Bucket has no impact on latency, RPS compared to Basic Litellm Proxy

MetricBasic Litellm ProxyLiteLLM Proxy with GCS Bucket Logging
RPS1133.21137.3
Median Latency (ms)140138

LangSmith logging

Using LangSmith has no impact on latency, RPS compared to Basic Litellm Proxy

MetricBasic Litellm ProxyLiteLLM Proxy with LangSmith
RPS1133.21135
Median Latency (ms)140132