Platform

One endpoint. Every request accounted for.

If your code talks to an OpenAI-style API, it already talks to MarsCompute. Keys, spending limits and metering are part of the platform, not something you add later.

Request access Read the quickstart

Request path

What happens on every call.

Five steps, in this order. Select one to read it.

The key is verified and matched to its one model and one member.

curl https://api.marscompute.ai/v1/chat/completions \
  -H "Authorization: Bearer $MARSCOMPUTE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-sea-lion-v4.5-27b",
    "stream": true,
    "messages": [
      {"role": "user", "content": "Say hello in Malay."}
    ]
  }'

import os
import requests

response = requests.post(
    "https://api.marscompute.ai/v1/chat/completions",
    headers={
        "Authorization": f"Bearer {os.environ['MARSCOMPUTE_API_KEY']}",
    },
    json={
        "model": "qwen-sea-lion-v4.5-27b",
        "messages": [
            {"role": "user", "content": "Say hello in Malay."}
        ],
    },
    timeout=120,
)
response.raise_for_status()
print(response.json()["choices"][0]["message"]["content"])

const response = await fetch(
  "https://api.marscompute.ai/v1/chat/completions",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer " + process.env.MARSCOMPUTE_API_KEY,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      model: "qwen-sea-lion-v4.5-27b",
      messages: [{ role: "user", content: "Say hello in Malay." }],
    }),
  },
);

const completion = await response.json();
console.log(completion.choices[0].message.content);

Example of a streamed reply Helo! Selamat datang. Apa khabar hari ini? usage: 19 input tokens, 12 output tokens

Built for production use

Control over cost and access, not only a model list.

Streaming

Responses arrive token by token over server-sent events, with token usage reported at the end of the stream.

Scoped API keys

Every key belongs to one member and one model. Revoke a key without touching the others.

Usage you can audit

Each request is metered by input and output tokens and listed with its cost. Export usage as CSV whenever you need it.

Spending limits

Set a monthly limit for your workspace. Cost is reserved before a request is sent, so the limit holds.

Team workspaces

Invite members to your workspace. Each member signs in with their own email and manages their own keys.

Safe retries

Send an Idempotency-Key header and a repeated request is not run or charged a second time.

Pricing

Pay for the tokens you use.

Prices are per one million tokens, in US dollars, and exclude applicable taxes.

Model	Model ID	Input / 1M	Output / 1M
Qwen-SEA-LION-v4.5-27B-IT	`qwen-sea-lion-v4.5-27b`	$0.60	$3.60
GPT-6 Astra	`gpt-6-astra`	$10.00	$50.00
Kimi K3	`k3`	Coming soon
GLM-5.3	`glm-5.3`	Coming soon

See how billing works

Bring the client you already use.

Request access and we will set up your workspace.

Request access See the models