LiteLLM

Route to 100+ LLM providers through a single LiteLLMModel interface. Use this when you want one code path for OpenAI, Anthropic, Azure, Vertex, Bedrock, and more.

Browse supported model names at the LiteLLM model hub.

For standalone invoke / stream usage, see LLM Models — LiteLLM.

Install

pip install --extra-index-url https://elsai-agents.elsai.ai/root/ elsai-agents==0.3.1
pip install --extra-index-url https://core-packages.elsai.ai/root/ "elsai-model[litellm]==2.1.0"

Setup

1. Set the provider API key for your chosen model (e.g. OPENAI_API_KEY, ANTHROPIC_API_KEY).
2. Pick a model ID from models.litellm.ai and export it:

export OPENAI_API_KEY=sk-...
export LITELLM_MODEL=gpt-4o-mini

Agent — basic

import os
from elsai import Agent
from elsai_model import LLM, Provider

model = LLM(
    provider=Provider.LITELLM,
    model=os.getenv("LITELLM_MODEL", "gpt-4o-mini"),
    api_key=os.environ["OPENAI_API_KEY"],
    params={"temperature": 0.2},
)
agent = Agent(
    model=model,
    system_prompt="You are a concise assistant.",
)
result = agent("Summarize what an AI agent is in two sentences.")
print(result)

Model ID examples

Provider	model_id example
OpenAI	gpt-4o-mini
Anthropic	anthropic/claude-3-5-sonnet
Azure OpenAI	azure/gpt-4o
Vertex AI	vertex_ai/gemini-1.5-pro

Set provider credentials via standard LiteLLM environment variables.