Integration
Use the LiteLLM library with Concentrate
Point LiteLLM completion calls at Concentrate to access models, track spend, and keep provider routes out of app code.
Library
litellm.completion
Route through https://api.concentrate.ai/v1.
Model
Provider/slug
Use Concentrate model names in requests.
Review
Logs
Debug failures and token spend by key.
New capabilities
What your team gains with Concentrate
01
Keep LiteLLM calls
Set api_base and api_key on LiteLLM instead of hosting a proxy yourself.
02
Reach more providers
Use Concentrate model routes from one Concentrate API key.
03
Track every completion
See spend, tokens, status, model, provider, and owner in one place.
Integration basics
Frequently asked questions
Do we still need a self-hosted LiteLLM proxy?
Not for library usage. Many teams point litellm.api_base at Concentrate and use a Concentrate API key instead of running LiteLLM Proxy for every environment.
Should we use api_base or base_url with LiteLLM?
Use api_base for litellm.completion and litellm.acompletion. The base_url parameter is ignored by LiteLLM and can send traffic to the wrong endpoint.
How is this different from migrating off LiteLLM Proxy?
This page is for teams keeping the LiteLLM library in application code. If you run LiteLLM Proxy as a service, see the LiteLLM migration guide for moving that gateway layer to Concentrate.