GPT-6 Astra represents a step-function change in model capability for interactive reasoning problems. It scores 66% on ARC-AGI-3 using our standard harness, and nearly 100% with a continuous conversation harness and custom compaction, at a cost of roughly $360 per game.
In fact,
Fable 5.1 is mind-blowingly good, but the limits on the Max 20x plan are a joke.
- You can easily burn through the 5 hour limit in about 20 minutes
- The "weekly" limit seems to be about 2x 5 hour limits' worth.
Seriously?!
So basically you can burn a week's worth of tokens
EMERGENCY PODCAST:
The recent paper on the reasoning heist went super viral.
It was also 100+ pages long and you didn't have time to read it.
The authors @iliaishacked and @kotekjedi_ml unpack the paper and discuss model distillation.
