OpenRouter added Space Bunny Alpha to its platform. It is a free, anonymous large model. It supports fast inference, strong coding, and native multimodal input. It offers a 1,000,000 token context window. It allows up to 524,288 output tokens. It supports adjustable reasoning effort. It accepts text, images, and video as input. It supports tool calling and structured JSON output. It is developed and operated by a third-party provider. The provider remains anonymous during this preview. OpenRouter routes requests to it directly. Prompts and completions may be retained but are not used for training.
What changed
No prior public version of this model existed.
What it unlocks
Using a free, anonymous large model with fast inference, strong coding, and native multimodal input support.
- 1,000,000 token context window
- 524,288 max output tokens
- 70 tokens per second throughput
- 2.10s latency
Sources