Ox Alpha
Former anonymous OpenRouter stealth model with 1M context, revealed as Z.ai GLM-5.3 Flash
Z.ai (Zhipu AI) has replaced this model with GLM-5.3-Flash.
Context window
1,048,576 (1M tokens)
Input / 1M tokens
Free during stealth period (ended)
Output / 1M tokens
Free during stealth period (ended)
Provider
Z.ai (Zhipu AI)
Ox Alpha was free on OpenRouter during its stealth period, which has ended; the stealth listing has no active endpoints. OpenRouter now says the model was developed and operated by Z.ai and was GLM-5.3 Flash, which is sold under its own name. · Data verified 2026-11-08
Ox Alpha was a reasoning model for coding, sustained agentic work, and production workloads, with a 1,048,576-token context window, multimodal input (text, image, video), tool calling, and structured output support. It appeared anonymously on OpenRouter on August 20, 2026 with free access during a stealth period. OpenRouter later revealed it as Z.ai's GLM-5.3 Flash, and the free stealth endpoint no longer serves requests, so use GLM-5.3 Flash instead.
Capability index
Relative estimates (0-100) to place this model against its peers, grounded in published benchmarks.
How to access it
No longer available: OpenRouter ended the free stealth period and the stealth/ox-alpha listing has no active endpoints. OpenRouter says the model was Z.ai's GLM-5.3 Flash, available as z-ai/glm-5.3-flash on OpenRouter and through Z.ai.
Strengths
- ✓1-million token context window for large codebases and long-horizon tasks
- ✓Multimodal input (text, image, video)
- ✓Strong early community performance on coding benchmarks (80% Pass@1 on DeepSWE)
- ✓Tool calling and structured JSON output support
Best for developers who...
When to choose it (and when not to)
Reach for Ox Alpha when...
- →Not for new work: the stealth endpoint is gone; use GLM-5.3 Flash, the released model behind Ox Alpha
Look elsewhere if...
- ✕Any new project (the stealth endpoint no longer serves requests)
- ✕Production workloads with sensitive data (stealth prompts and completions were retained by the provider)
How to use it
- ›Include full codebase context - the 1M window is designed for repository-scale work
- ›Use tool_choice to direct the model to function calling for structured tasks
- ›Test real engineering workflows (bug fixes, feature additions) rather than simple chat prompts
Quickstart
Pythonimport openai
client = openai.OpenAI(
api_key="your-openrouter-key",
base_url="https://openrouter.ai/api/v1"
)
response = client.chat.completions.create(
model="stealth/ox-alpha",
messages=[{"role": "user", "content": "Fix the bug in this code..."}]
)
print(response.choices[0].message.content)Historical example: the stealth/ox-alpha endpoint no longer serves requests. Call GLM-5.3 Flash (z-ai/glm-5.3-flash on OpenRouter) instead.
API model id: stealth/ox-alpha
Benchmarks
| Benchmark | Score | Notes |
|---|---|---|
| DeepSWE (community test subset) | 80% Pass@1 | User-run test on 10 tasks, unaudited; outperformed GPT-5.6-SOL (~52%) and Claude Opus 5 (~65%) on tested subset but sample size is small and statistical noise is high. |
Compare Ox Alpha
Compare Ox Alpha with any other model
Build a comparison →All model comparisons →