From 50ebbc7a7ad6cb738bbcd5bcb562ff5ba8cc0c6a Mon Sep 17 00:00:00 2001 From: Alexsander Hamir Date: Sat, 10 Jan 2026 16:23:14 -0800 Subject: [PATCH] docs: improve performance section explanation with technical details Add details about O(1) dictionary lookup optimization and explain why this has high impact (runs in client decorator on every proxy request). --- docs/my-website/release_notes/v1.80.14/index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/my-website/release_notes/v1.80.14/index.md b/docs/my-website/release_notes/v1.80.14/index.md index 06861047395..5ecb3c13e4d 100644 --- a/docs/my-website/release_notes/v1.80.14/index.md +++ b/docs/my-website/release_notes/v1.80.14/index.md @@ -61,7 +61,7 @@ pip install litellm==1.80.14 ## Performance - 50% Overhead Reduction -LiteLLM now sends 2.5× more requests to LLM providers by reducing wrapper function overhead across all endpoints. +LiteLLM now sends 2.5× more requests to LLM providers by replacing sequential if/elif chains with O(1) dictionary lookups for provider configuration resolution (92.7% faster). This optimization has a high impact because it runs inside the client decorator, which is invoked on every HTTP request to the proxy server. ### Before