SPACE BUNNY ALPHA API · OPENAI-COMPATIBLE

Space Bunny API

Call MiniMax M3.1 Flash from your code with the OpenAI SDK you already use. Point the base URL at Space Bunny, set the model to minimax-m3.1-flash and pay 1 credit per request, shared with your Space Bunny chat balance.

Get an API key See credit plans

Base URL: https://spacebunny.pro/api/v1
Endpoint: POST /chat/completions
Model: minimax-m3.1-flash
Auth: Bearer sk_spacebunny_…

QUICKSTART

Your first Space Bunny API request

curl https://spacebunny.pro/api/v1/chat/completions \
  -H "Authorization: Bearer $SPACEBUNNY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-m3.1-flash",
    "reasoning_effort": "medium",
    "messages": [{ "role": "user", "content": "Write a haiku about orbital mechanics." }]
  }'

Your Space Bunny API keys

Keys belong to your Space Bunny account and use its credit balance. Keep them on your server and revoke any key you no longer use.

PARAMETERS

Request fields

model"minimax-m3.1-flash"

Optional; the earlier name "space-bunny-alpha" still works.

messagesarray

system, developer, user, assistant and tool roles. User content may mix text and image_url parts.

reasoning_effortlow · medium · high

Default medium. Also accepted as reasoning.effort. Set reasoning.exclude to hide reasoning text.

streamboolean

Server-sent events in the OpenAI chunk format, ending with [DONE]. Reasoning arrives in delta.reasoning.

max_tokens1–32,768

Default 16,000. Reasoning tokens count toward it. max_completion_tokens also works.

tools / tool_choicefunction tools · "auto" | "none"

Up to 64 function tools.

response_formatjson_object · json_schema

Structured JSON output.

temperature / top_p0–2 / 0–1

Optional sampling controls.

LIMITS & BILLING

One request, one credit

Price: 1 credit per request
Input: Up to 512,000 characters (~128K tokens)
Images: Up to 4 per request, 3 MB each
Rate: 30 requests / minute per key
Concurrency: 3 running requests per account
Body: Up to 4 MB JSON

The credit is held when a request starts. If the model fails, times out, or the stream disconnects before any answer text or tool call is sent, the credit is returned. Check your balance with GET /api/v1/credits.

ERRORS

Error responses

Errors use the OpenAI shape { "error": { "message", "type", "code" } }. Every response carries an X-Request-Id header to quote when contacting support.

401

Missing, invalid or revoked API key.

402

Not enough credits. Nothing is charged.

400 / 404 / 413

Invalid request, unknown model or too large. Nothing is charged.

429

Per-key rate limit or account concurrency limit reached.

502 / 503

The model failed or is at capacity. The credit is returned.

Build with MiniMax M3.1 Flash

New accounts get 20 free credits to try the chat and the API.

Create an API key

Requests are processed by hosted providers of MiniMax M3.1 Flash. Do not send secrets or personal data. Space Bunny is not affiliated with MiniMax or its hosting providers.