Three weeks ago, I wrote about losing access to Fable 5 after 72 hours. The government pulled it offline before most developers even got a chance to try it. That article hit a nerve — turns out a lot of people felt the same frustration.
Here's what I didn't write about at the time: those 72 hours changed how I think about what an AI agent should be capable of. And it directly shaped what we built next.
What Fable Actually Proved
Forget the benchmarks for a second. Yes, Fable 5 scored 80.3% on SWE-Bench Pro. Yes, it hit 29.3% on FrontierCode Diamond — five times the nearest competitor. Those numbers matter, but they're not the point.
The point is what it felt like to work with. For three days, I had an agent that could:
