Back to feed
News Story
InfoQ
1 sources

Mysterious Model 'Ox Alpha' Surprises on OpenRouter, Outperforming Fable 5? Speculation Points to Zhipu or Xiaomi

OpenRouter has released a mysterious model named Ox Alpha, featuring a 1 million token context window, support for text, image, and video inputs, and is currently free. The developer behind the model remains anonymous, sparking speculation about its origin. The model targets coding and agent tasks, but benchmark results have not been disclosed.

SynthePulse Insight · AI deep readingMembers

Mysterious Model Ox Alpha Surfaces on OpenRouter: Free Million-Token Context, Identity Unknown

Version 1 · 1 source

OpenRouter released an anonymous model, Ox Alpha, with a ninja emoji, offering a free million-token context, sparking community speculation about its origin, but performance and identity remain unconfirmed.

  • OpenRouter anonymously released the model Ox Alpha, offering a free 1 million token context, supporting text, image, and video inputs, with text-only outputs.
  • Ox Alpha supports tool calling and JSON structured outputs, but not strict JSON Schema, positioning it for long-horizon agent tasks.
  • OpenCode offers 100 trillion tokens per day for free testing for one week, prompting a large-scale real-user stress test.
Open section navigationAnonymous Release and Core Features

Anonymous Release and Core Features

On August 23, OpenRouter, without a launch event or technical report, released a model codenamed Ox Alpha with a ninja emoji, with the full identifier stealth/ox-alpha. The model is described as a frontier model for efficient coding, long-horizon agent tasks, and real production environments, featuring a context window of over 1 million tokens, a maximum output of 130,000 tokens, and support for text, image, and video inputs, but text-only outputs.

Ox Alpha supports tool calling, tool_choice, and JSON structured outputs, but not strict JSON Schema constraints. These features indicate it is not a simple chat or code completion model, but an execution model designed for coding agents, emphasizing sustained agentic work.

OpenRouter has not published any benchmark scores, such as SWE-bench or Terminal-Bench, so there is no public evidence of whether its performance reaches the top tier. Model parameters, training data, architecture, and development team are all confidential, and OpenRouter acts only as a router, with the model provided by an anonymous third-party vendor.

Free for now

Read the full analysis

3 more sections of analysis, plus the full takeaway

Loading

Credibility boundary

This article's information primarily comes from InfoQ reporting, including community tests and speculations, none of which have been officially confirmed. Performance data comes from user tests, with small samples and inconsistent criteria, and cannot be taken as definitive. Identity speculations lack conclusive evidence.

Primary report

InfoQ

Primary source