space bunny
Space Bunny is an anonymous 1M-context reasoning model for text, image and video prompts, adjustable effort, JSON output, and approved tool workflows.
Screenshots
Overview
Product overview
Space Bunny is an anonymous preview reasoning model with a one-million-token context window. It gives developers a playground and an OpenAI-compatible API for coding, research, visual understanding, structured responses, and agent workflows.
Core capabilities
- Work across long documents, repositories, and extended conversations with 1M tokens of context.
- Accept text, image URLs, and compatible video URLs, then return a text answer.
- Select low, medium, high, xhigh, or max reasoning effort to trade speed for depth.
- Ask for JSON objects when the next step needs predictable data.
- Let the model request approved functions while the application validates permissions, arguments, and side effects.
- Use standard chat-completions fields, usage metadata, and server-side API keys.
Workflow
- Try a real prompt in the Playground with low reasoning effort.
- Create a Space Bunny key and keep it in a server-side environment variable.
- Send the model ID, ordered messages, and only the multimodal or tool controls needed for the task.
- Read the assistant message and usage, parse JSON after validation, or approve a tool call inside your own application.
Who it fits
Space Bunny suits developers reviewing a large codebase, teams examining screenshots or diagrams, and agent builders who need long context with explicit tool-safety boundaries. It also supports research and incident-review prompts that benefit from a deeper reasoning setting.
Plans
The playground preview is free. The pricing page lists one-time credit packs with no subscription, no auto-renewal, and no expiry: Starter ($9.90 / 100,000 credits), Pro ($99 / 1,000,000 credits), and Enterprise ($999 / 11,000,000 credits including 10% extra).
