Skip to main content
Use this guide when your LlamaIndex app should route generation or embeddings through AI Stats.

​
Install

pip install llama-index llama-index-llms-openai-like llama-index-embeddings-openai-like

​
Configure an OpenAI-compatible LLM

import os
from llama_index.llms.openai_like import OpenAILike

llm = OpenAILike(
    model="openai/gpt-5-nano",
    api_key=os.environ["AI_STATS_API_KEY"],
    api_base="https://api.phaseo.app/v1",
    is_chat_model=True,
)

response = llm.complete("Write one sentence about retrieval quality.")
print(response)

​
Use in an index

from llama_index.core import Settings

Settings.llm = llm
Then build or query your index as usual.

​
Notes

  • Use OpenAILike for third-party OpenAI-compatible APIs.
  • Keep generation and embedding model ids explicit. Do not assume LlamaIndex defaults point at AI Stats.
  • If you use AI Stats embeddings, set the embedding model with an AI Stats model id and the same base URL.
Last modified on June 11, 2026