Space Bunny Alpha on OpenRouter

OpenRouter was where Space Bunny first appeared, and it is still the easiest way to call Space Bunny from code. OpenRouter exposes an OpenAI-compatible API, so any client that talks to OpenAI can talk to Space Bunny after you change two settings: the base URL and the model name.

The quickest test is a curl call. Low effort keeps the first Space Bunny response fast:

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "stealth/space-bunny-alpha",
    "reasoning": { "effort": "low" },
    "messages": [
      { "role": "user", "content": "Write a Python function that dedupes a list while keeping order." }
    ]
  }'

The same request in Python, using the official OpenAI SDK pointed at OpenRouter:

from openai import OpenAI

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="YOUR_OPENROUTER_KEY",
)

resp = client.chat.completions.create(
    model="stealth/space-bunny-alpha",
    messages=[{"role": "user", "content": "Explain this stack trace and suggest a fix: ..."}],
    extra_body={"reasoning": {"effort": "medium"}},
)
print(resp.choices[0].message.content)

OpenRouter accepts the reasoning object in the request body. With the OpenAI SDK you pass it through extra_body. The OpenRouter response includes a reasoning field when you ask for it, so you can inspect how Space Bunny worked through the problem.

Space Bunny accepts images and video through OpenRouter's standard multimodal message format. A screenshot-to-code request looks like this:

resp = client.chat.completions.create(
    model="stealth/space-bunny-alpha",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "Rebuild this screen as a single HTML file with Tailwind."},
            {"type": "image_url", "image_url": {"url": "https://example.com/screenshot.png"}},
        ],
    }],
    extra_body={"reasoning": {"effort": "high"}},
)

Video uses the same content array with a video part. Keep clips short at first, since video tokens add up quickly even in a 1M window. Space Bunny outputs text only, so it can describe or rebuild what it sees but cannot generate images.

• 404, no endpoints found: your privacy settings exclude providers that retain prompts. Allow them for the Space Bunny key.
• 429, rate limited: you hit the free limit of about 20 requests per minute or the daily cap. Slow down or add credits to raise the cap.
• Empty or cut-off answers: raise max_tokens. Reasoning tokens count toward the budget, so a small limit can be spent entirely on thinking.
• Language drift: a few users saw Chinese appear mid-answer. Add "Respond only in English" to the system prompt.
• Short outages: the Space Bunny endpoint went down briefly on launch day. Retries with backoff and a fallback model handle this.