LlamaIndex

LlamaIndex can use OdiRouter through OpenAI-compatible LLM clients. Set the API key, API base, and OdiRouter model ID.

Requirements

Requirement	Value
Python	3.10+
Package	LlamaIndex OpenAI LLM integration
API base	https://www.odirouter.ai/v1
API Key	OdiRouter API key
Model	An OdiRouter LLM model ID

Install

pip install llama-index llama-index-llms-openai

Configure LLM

import os

from llama_index.llms.openai import OpenAI

llm = OpenAI(
    model="gpt-5.5",
    api_key=os.environ["ODIROUTER_API_KEY"],
    api_base="https://www.odirouter.ai/v1",
)

Complete Text

response = llm.complete("Reply with a one-line LlamaIndex integration test.")

print(response.text)

Use With Settings

from llama_index.core import Settings

Settings.llm = llm

Use separate embedding settings unless your selected OdiRouter model and LlamaIndex integration explicitly support embeddings.

Troubleshooting

Symptom	Fix
Requests go to OpenAI	Confirm api_base is set to https://www.odirouter.ai/v1
401 response	Set ODIROUTER_API_KEY in the server environment
Model not found	Use an exact model ID from /models
RAG app fails during embedding	Configure embeddings separately from the OdiRouter LLM client