LlamaIndex + Evrona
LlamaIndex accepts an OpenAI-compatible address right in the LLM constructor — no extra package is required.
What to prepare
No key yet? Create an account — keys are created in the dashboard under “API keys”, and the full value is shown only once. Copy the Model ID from the catalog in full, including the prefix before “/”.
Connection steps
Install the llama-index-llms-openai package.
Build the LLM with api_base, your key and a Model ID.
Hand it to your index or query engine in the usual way.
# pip install llama-index-llms-openai
from llama_index.llms.openai import OpenAI
llm = OpenAI(
model="google/gemini-3.7-flash",
api_base="https://evrona.ru/v1",
api_key="ev-YOUR_KEY",
is_chat_model=True,
context_window=128000,
)
print(llm.complete("Hello"))is_chat_model=True has to be set: without it LlamaIndex goes off to the legacy /v1/completions path, which the gateway does not serve. Set context_window to the window of the model you picked from the catalog.
Function calling is supported, so agentic modes work. Strict JSON mode (response_format) and the retired functions format are turned down with an unsupported_parameter error — no parameter is lost in silence. Image input is taken by models marked "Understands images" in the catalog: a content part of type image_url with a data:image/png;base64,… URL (PNG, JPEG or WebP, up to 5 MB each, up to 50 per request). Remote URLs are not fetched, and a model without that mark returns unsupported_capability — both errors name the reason outright.