Quickstart

belweave infer uses the OpenAI Chat Completions format. Existing code needs two changes: the base URL and the key.

Get a key

Email info@belweave.com with your name and a short description of your software or engineering work. We verify your identity before issuing a key. Limits are arranged per access when your key is issued.

Email info@belweave.com

For immediate use or interest, email info@belweave.com.

Base URL and auth

Authentication

Base URL:  https://api.infer.belweave.ai/v1
Header:    Authorization: Bearer $BELWEAVE_API_KEY

Keep the key in an environment variable, not in source control.

First request

example/v1/chat/completions

cURL

curl https://api.infer.belweave.ai/v1/chat/completions \
  -H "Authorization: Bearer $BELWEAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
    "messages": [
      {"role": "system", "content": "You are a software engineering assistant."},
      {"role": "user", "content": "Explain how to make an API request safe to retry."}
    ]
  }'

Python

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.infer.belweave.ai/v1",
    api_key=os.environ["BELWEAVE_API_KEY"],
)

resp = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V4.1-Flash-1017",
    messages=[
        {"role": "system", "content": "You are a software engineering assistant."},
        {"role": "user", "content": "Explain how to make an API request safe to retry."},
    ],
)
print(resp.choices[0].message.content)

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.infer.belweave.ai/v1",
  apiKey: process.env.BELWEAVE_API_KEY,
});

const resp = await client.chat.completions.create({
  model: "deepseek-ai/DeepSeek-V4.1-Flash-1017",
  messages: [{ role: "user", content: "Write a test for retrying a failed API request." }],
});
console.log(resp.choices[0].message.content);

Streaming

Set stream: true to receive tokens as server-sent events (SSE).

example/v1/chat/completions

Python

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.infer.belweave.ai/v1",
    api_key=os.environ["BELWEAVE_API_KEY"],
)

stream = client.chat.completions.create(
    stream=True,
    model="deepseek-ai/DeepSeek-V4.1-Flash-1017",
    messages=[
        {"role": "system", "content": "You are a software engineering assistant."},
        {"role": "user", "content": "Explain how to make an API request safe to retry."},
    ],
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

Supported features

Feature	Use
Chat completions	Send messages to `POST /v1/chat/completions`
Streaming	Set `stream: true` for server-sent events
Tool calling	Get tool calls for your application to handle
Reasoning output	Receive reasoning alongside the answer

Errors

Code	Meaning	What to do
401	Missing or invalid key	Check the Authorization header
403	Key suspended	Email info@belweave.com
429	Rate limited	Back off and retry. Limits are arranged when your key is issued.
503	Service temporarily unavailable	Check Status and retry

Retries

Retry temporary errors with backoff and make long jobs resumable. Check Status for service information.