Skip to main content
LiteLLM is a lightweight Python library for calling 100+ LLM APIs in the same format. You can point it at Lightweight using CLI flags or environment variables.

​
CLI Usage

litellm --model openai/claude-sonnet-4.5 \
        --api_base https://api.lightweight.one/v1 \
        --api_key lw_sk_your-key-here
LiteLLM requires the openai/ prefix when using OpenAI-compatible providers.

​
Python Usage

import litellm

response = litellm.completion(
    model="openai/gpt-5.4",
    api_base="https://api.lightweight.one/v1",
    api_key="lw_sk_your-key-here",
    messages=[
        {"role": "user", "content": "What is the capital of France?"}
    ]
)

print(response.choices[0].message.content)

​
Environment Variables

Alternatively, set environment variables so you don’t need to pass credentials on every call: 
export OPENAI_API_BASE="https://api.lightweight.one/v1"
export OPENAI_API_KEY="lw_sk_your-key-here"
Then call LiteLLM without explicit credentials: 
response = litellm.completion(
    model="openai/claude-sonnet-4.5",
    messages=[{"role": "user", "content": "Hello!"}]
)
LiteLLM supports streaming, function calling, and all other OpenAI-compatible features through Lightweight. See the full Models catalog.