OpenAI Python SDK with Meridian

Install

pip install openai

Quick start

from openai import OpenAI

client = OpenAI(
    api_key="nim_live_***",
    base_url="https://meridian.getnimbus.net/api/v1"
)

resp = client.chat.completions.create(
    model="azure-swc/gpt-4.1",
    messages=[{"role": "user", "content": "hello"}]
)

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

Streaming

Set stream=True and iterate chunks — identical to native OpenAI behavior.

for chunk in client.chat.completions.create(
    model="azure-swc/gpt-4.1",
    messages=[{"role": "user", "content": "Tell me a story"}],
    stream=True
):
    print(chunk.choices[0].delta.content or "", end="", flush=True)

Tool calls

Function calling works out of the box. Define tools, inspect tool_calls on the response, execute, and return results in a follow-up message.

tools = [{
    "type": "function",
    "function": {
        "name": "get_weather",
        "description": "Get current temperature for a city",
        "parameters": {
            "type": "object",
            "properties": {
                "city": {"type": "string"}
            },
            "required": ["city"]
        }
    }
}]

resp = client.chat.completions.create(
    model="azure-swc/gpt-4.1",
    messages=[{"role": "user", "content": "What's the weather in London?"}],
    tools=tools
)

tool_call = resp.choices[0].message.tool_calls[0]
print(tool_call.function.name, tool_call.function.arguments)

Vision

Pass image URLs or base64-encoded images in the content array — same multimodal format.

resp = client.chat.completions.create(
    model="azure-swc/gpt-4.1",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "Describe this image"},
            {
                "type": "image_url",
                "image_url": {"url": "https://example.com/photo.jpg"}
            }
        ]
    }]
)

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

JSON mode

Guarantee structured output with response_format={"type": "json_object"}.

resp = client.chat.completions.create(
    model="azure-swc/gpt-4.1",
    messages=[{"role": "user", "content": "List three colors as JSON"}],
    response_format={"type": "json_object"}
)

import json
data = json.loads(resp.choices[0].message.content)
print(data)