OpenAI and Anthropic SDK compatibility

Meridian implements the OpenAI Chat Completions API and the Anthropic Messages API exactly. Drop in your existing SDK with a new base URL and key. No client rewrites.

OpenAI Python SDK

Install openai and set the base URL to Meridian. Your code stays the same.

from openai import OpenAI

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

resp = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

OpenAI TypeScript SDK

Use the official client with the Meridian endpoint. Models and payload format are compatible with Chat Completions.

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "nim_live_...",
  baseURL: "https://meridian.getnimbus.net/api/v1",
});

const response = await client.chat.completions.create({
  model: "auto",
  messages: [{ role: "user", content: "Hello" }],
});

console.log(response.choices[0].message?.content);

Anthropic Python SDK

Point the Anthropic client at Meridian and keep your message payloads unchanged.

from anthropic import Anthropic

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

response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=128,
    messages=[{"role": "user", "content": "Hello"}],
)

print(response.content[0].text)

Anthropic TypeScript SDK

Swap the base URL and key. Meridian forwards the Messages API while keeping the payload schema intact.

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "nim_live_...",
  baseURL: "https://meridian.getnimbus.net/api/v1",
});

const result = await client.messages.create({
  model: "claude-3-5-sonnet-20241022",
  max_tokens: 128,
  messages: [{ role: "user", content: "Hello" }],
});

console.log(result.content[0].text);

Vercel AI SDK

Use the OpenAI adapter with the Meridian base URL. The adapter handles the transport so you keep the same streaming and tool abstractions.

import { createOpenAI } from "@ai-sdk/openai";

const meridian = createOpenAI({
  apiKey: process.env.MERIDIAN_API_KEY,
  baseURL: "https://meridian.getnimbus.net/api/v1",
});

const result = await meridian.chat("auto").doGenerate({
  prompt: [{ role: "user", content: [{ type: "text", text: "Hello" }] }],
});

Drop-in environment variable trick

If you cannot change SDK initialization code, override the base URL and key at the environment level. Most SDKs read these variables on startup.

export OPENAI_API_KEY="nim_live_..."
export OPENAI_BASE_URL="https://meridian.getnimbus.net/api/v1"

export ANTHROPIC_API_KEY="nim_live_..."
export ANTHROPIC_BASE_URL="https://meridian.getnimbus.net/api/v1"