LiteLLM CLI Cheatsheet - Unified LLM Gateway & Proxy Server

All LiteLLM unified-gateway commands in one place — from one-line proxy to config-driven mode, multi-provider routing, cost/rate-limit — organized by scenario. Copy and go.

AI CLI·25 commands·Last updated 2026-08-23

Install 5

uv tool install 'litellm[proxy]'

Recommended: install proxy build via uv

pip install "litellm[proxy]"

Alternative: pip install proxy build

litellm --version

Show installed version

litellm --help

Show all subcommands

python -c "import litellm; print(litellm.__version__)"

Confirm the Python package loads

Start Proxy (one-liner) 5

litellm --model gpt-3.5-turbo

OpenAI-compatible proxy (default :4000)

litellm --model anthropic/claude-...

anthropic/ prefix routes to Claude

litellm --model ollama/

ollama/ prefix uses local Ollama

litellm --model bedrock/anthropic.claude-v2

bedrock/ prefix uses AWS Bedrock

litellm --model vllm/<model>

vllm/ prefix uses local vLLM

Config-driven (prod) 5

litellm --config config.yaml

Launch from a config (model list/keys/routing)

litellm --config config.yaml --detailed_debug

Verbose debug logging for troubleshooting

litellm --test

Self-test: send one request to the proxy

litellm --test <model>

Connectivity test for a specific model

export OPENAI_API_KEY=...

Set the backend provider key

Routing & Ops 5

openai_api_base=http://0.0.0.0:4000

Point clients at the LiteLLM proxy

curl http://0.0.0.0:4000/v1/models

List models exposed by the proxy

litellm --model <provider>/ --api_base <url>

Point at a custom OpenAI-compatible endpoint

litellm --drop_params

Drop unsupported params to improve compatibility

litellm --telemetry False

Disable telemetry (privacy/offline)

Tips 5

One-liner suits local validation

Quickly try different providers

Use --config in prod, not one-liners

Centralize keys/routing/budgets

provider/ prefix is the key

openai/ anthropic/ ollama/ decide routing

--test after config changes

Avoid discovering breakage in prod

Proxy unifies the OpenAI protocol

Old OpenAI clients just change base_url

Tips

LiteLLM uses the `provider/` prefix to route 100+ models to one OpenAI-compatible interface — the de-facto gateway for multi-provider fallback, cost, and rate-limiting.
For quick local validation use a one-liner `litellm --model <provider/model>`; in production always use `--config config.yaml` to centralize keys, routing, and budgets.
The proxy listens on :4000 by default; legacy OpenAI clients switch providers by just changing `base_url` — near-zero migration cost.
After config changes, run `litellm --test` to self-check connectivity before shipping.

Official References

Commands are compiled from the official docs below. Click to verify the latest usage.

Maintained by LaoHand

Publicly updated on Aug 23, 2026, continuously proofread against official docs.

Contact Us

Wrong command or description? Send us corrections, business inquiries or product feedback by email.

Edit this page on GitHub

Edit the source file directly and open a PR to improve this cheatsheet.

Edit this page on GitHub Report an issue