MixerBox Cloud

Docs

Build, then ship.

MixerBox Cloud is OpenAI-compatible. Point your client at our inference endpoint and start sending requests — 30 seconds to ship.

Quick start

  1. 1. Sign in at /dashboard with your MixerBox ID.
  2. 2. Create an API key from /dashboard. Treat it like a password — don't share or commit it.
  3. 3. Top up at least $5 of credit (Stripe).
  4. 4. Point your OpenAI client at https://cloud.mixerbox.com/v1.

Authentication

All requests require a Bearer token in the Authorization header. Get your key from the dashboard.

Header 
Authorization: Bearer sk-mb-...

Chat completions

Send a list of messages to our inference endpoint and receive a model-generated reply. Identical to the OpenAI /v1/chat/completions spec.

Request

curl 
curl -X POST https://cloud.mixerbox.com/v1/chat/completions \
  -H "Authorization: Bearer sk-mb-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mixerbox/kimi-k2.5",
    "messages": [
      {"role": "user", "content": "Explain quantum computing in one paragraph."}
    ]
  }'

Response

JSON 
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "model": "mixerbox/kimi-k2.5",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Quantum computing harnesses..."
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 14,
    "completion_tokens": 85,
    "total_tokens": 99
  }
}

Streaming

Add "stream": true to receive Server-Sent Events (SSE). The final chunk includes usage.

curl 
curl -X POST https://cloud.mixerbox.com/v1/chat/completions \
  -H "Authorization: Bearer sk-mb-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mixerbox/kimi-k2.5",
    "messages": [
      {"role": "user", "content": "Hello"}
    ],
    "stream": true
  }'

SSE output

SSE 
data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"delta":{"role":"assistant"},"index":0}]}

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"delta":{"content":"Hello"},"index":0}]}

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"delta":{"content":"!"},"index":0,"finish_reason":"stop"}],"usage":{"prompt_tokens":9,"completion_tokens":2,"total_tokens":11}}

data: [DONE]

Python

Works with the official openai package. Just change the base_url.

Python 
from openai import OpenAI

client = OpenAI(
    base_url="https://cloud.mixerbox.com/v1",
    api_key="sk-mb-...",
)

response = client.chat.completions.create(
    model="mixerbox/kimi-k2.5",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
)

for chunk in response:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Node.js / TypeScript

Works with the official openai package. Just change the baseURL.

TypeScript 
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://cloud.mixerbox.com/v1",
  apiKey: "sk-mb-...",
});

const stream = await client.chat.completions.create({
  model: "mixerbox/kimi-k2.5",
  messages: [{ role: "user", content: "Hello!" }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || "");
}

API reference

Endpoints exposed under /v1:

Method Path Purpose
POST /v1/chat/completions Inference endpoint for chat-style generation (streaming or non-streaming)
GET /v1/models List models hosted on the Service with pricing and capabilities

Available models

Models hosted on our serverless inference. Prices per 1M tokens, in USD.

Model ID Context Input $/1M Output $/1M
mixerbox/kimi-k2.5 256K $0.39 $1.89
mixerbox/minimax-m2.5 196K $0.14 $0.89
mixerbox/glm-5 200K $0.59 $1.91
mixerbox/glm-4.7 203K $0.39 $1.74
mixerbox/qwen3-coder 256K $0.21 $0.99
mixerbox/nemotron-30b 256K $0.04 $0.19
mixerbox/nemotron-super 256K $0.08 $0.44
mixerbox/kimi-k2-thinking 256K $0.59 $2.49
mixerbox/gemma-3-4b 128K $0.03 $0.07
mixerbox/nemotron-9b 128K $0.03 $0.15
mixerbox/nemotron-12b-vl 128K $0.17 $0.50
mixerbox/gpt-oss-safeguard 128K $0.13 $0.50

Error codes

Standard HTTP status codes. Error responses follow the OpenAI error format.

Code Meaning
401 Missing or invalid API key.
402 Insufficient balance. Top up at /dashboard.
404 Model not found. Check /v1/models for available IDs.
429 Rate limited. Back off and retry with exponential delay.
500 Internal error. Retry once; if persistent, contact cloud@mixerbox.com.
503 Model temporarily unavailable. Retry or try a different model.

SDKs

No bespoke SDK needed. The official OpenAI client libraries work out of the box — just override the base URL.

OpenAI compatible

Our inference service implements the OpenAI spec exactly — every endpoint shape, field, and error code matches. Use it to generate client SDKs, import into Postman / Insomnia, or feed an AI agent. Just point the base URL at https://cloud.mixerbox.com/v1.