AI Summary
5 min readThe Mystery Model That Took Over OpenRouter
On August 20th, a stealth model appeared on OpenRouter—a platform that hosts various AI models for public use. It had no name, no branding, and no announcement. But it came with an astonishing offer: 100 trillion tokens of free usage, a million-token context window, and benchmarks that rivaled frontier models. The cost of serving that much inference, the hosts calculate, would be roughly $15 to $20 million per day if billed through a provider like OpenAI or Anthropic. The immediate question was obvious: who on earth is giving away that much compute for nothing?
The model was initially called "Zero X Alpha," but users quickly noticed the name was actually "Ox Alpha"—a small detail that hinted at Chinese zodiac symbolism. Over the next six days, something even stranger happened. The model appeared to improve day over day. Tasks it handled poorly on day one it handled significantly better by day six. This sparked speculation about "continual learning," the ability for a model to keep learning from new data without forgetting previous knowledge—a capability no major frontier lab has publicly achieved.
The Clues That Identified Ox Alpha
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Limitless Podcast
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (00:01) **The Mystery of Ox Alpha** - A stealth model appears on OpenRouter with unbelievable free offerings, sparking a week-long investigation into its identity.
- 2 (00:40) **Clues to the Model's Origin** - The hosts begin sleuthing, using tokenizers, political censorship, and naming conventions to narrow down the creator.
- 3 (03:13) **Viral Launch and Benchmark Performance** - Ox Alpha becomes the most popular model on OpenRouter overnight, dethroning DeepSeek by a factor of two.
- 4 (07:05) **The Flash Model Advantage** - The hosts explain why this model is so cheap to serve and what it means for the AI landscape.
- 5 (08:18) **Elon Musk vs. The Founder of Zhipu** - A recent exchange foreshadows this exact moment.
- 6 (09:02) **Debunking the Continual Learning Myth** - The hosts push back on the viral narrative that the model self-improves every day.
- 7 (10:00) **The Pepsi Challenge of AI** - A blind taste test reveals that users often can't tell the difference between Ox Alpha and frontier models.
+ Full timestamped outline available in the app
Show Notes
We discuss the World Humanoid Olympics in China, where humanoid robots competed in human-style events and set new performance marks. We also cover autonomy, reward-based training, and the comparison between China and the U.S. in robotics and AI.
------
🔒 Check Out Our Sponsor: LEDGER AGENT STACK 🔒
https://developers.ledger.com/docs/ai-tools/overview/?utm_source=Audio&utm_medium=Podcasts&utm_campaign=Limitless
------
🌌 LIMITLESS HQ ⬇️
NEWSLETTER: https://limitlessft.substack.com/
FOLLOW ON X: https://x.com/LimitlessFT
SPOTIFY: https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQ
APPLE: https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890
RSS FEED: https://limitlessft.substack.com/
------
TIMESTAMPS
0:00 Mystery Model Appears
1:30 Stealth Launch On OpenRouter
4:31 Chasing The Model’s Origin
6:09 Benchmarking Ox Alpha
6:59 Flash Model Theory
10:00 Blind Taste Test Buzz
12:11 Why Continual Learning Matters
14:58 Data, Subsidies, And Strategy
18:47 China’s Copycat Advantage
20:21 Mystery Solved For Now
21:11 More Models Incoming
------
RESOURCES
Josh: https://x.com/JoshKale
Ejaaz: https://x.com/cryptopunk7213
------
Not financial or tax advice. See our investment disclosures here:
https://www.bankless.com/disclosures
Josh works with Anthropic as a contractor. All views expressed are his own and do not represent Anthropic, its leadership, or its affiliates. Nothing in this episode is investment advice.
More from this podcast
Limitless Podcast →