---

license: other

language:

- en

tags:

- placeholder

- quantization

- gguf

- ox-alpha

- stealth-model

pipeline_tag: text-generation

---

Ox Alpha — Quantized GGUF · Placeholder

> Follow this repository to be notified if an official checkpoint becomes available for legal redistribution and quantization.

> 🐦 Follow @procrastiness on X for Ox Alpha GGUF release updates.

Status: no weights yet

This is a community placeholder, not a model release. There are currently no public Ox Alpha weights here, and this repository is not affiliated with the model developer, OpenRouter, OpenCode, Z.ai, or Hugging Face.

Ox Alpha appeared on OpenRouter on 20 August 2026 as stealth/ox-alpha. The hosted preview advertises a 1,048,576-token context window, up to 131,072 output tokens, multimodal input (text, images, and video), tool calling, structured output, and mandatory reasoning.

The developer and architecture have not been officially disclosed. Community reports suggesting a relationship to the GLM family remain unverified speculation.

Planned artifacts — only if source weights and licensing permit

• GGUF: Q4_K_M, Q5_K_M, Q6_K, Q8_0
• Quantization notes and reproducible commands
• Basic perplexity and coding sanity checks
• Checksums and source-revision provenance

Why follow?

If an official, redistributable checkpoint is released, this page will be updated with the source model, exact license, quantization method, checksums, and measured quality tradeoffs.

⭐ Like the repository to signal interest.

🔔 Follow this repository and @procrastiness for release updates.

Important

Do not download files claiming to be Ox Alpha weights unless their provenance and license can be verified. A hosted API model cannot be legitimately converted into GGUF without access to an authorized source checkpoint.

Known access

• Hosted model ID: stealth/ox-alpha
• Availability, price, limits, retention policy, and specifications may change during the preview.

Changelog

• 2026-08-22: Placeholder created; no weights uploaded.