Astra's 99.9% Score Came From the Harness, Not the Model
OpenAI's ARC-AGI-3 results sheet lists 62.7% and 99.9% side by side for the same model. Tracing why one model ends up with two scores reveals who actually deserves credit for this result.
350
ARC Prize
Runs ARC-AGI, an abstract reasoning benchmark.
Current rank (1M)
66
- Change
- Prev.
- –
Rank over the last 6 days
2026.09.08 – 2026.09.13 · High 65 · Low 74 · Now 66
ARC Prize · Related stories
AI models talk faster without words, thanks to a Fields Medalist's bridge
When we connect two models, we've always had one write text for the other to read.
200
Prime Agent technical report shows ARC-AGI-3 score jump from 30% to 95.5%
Prime Intellect says the gain came from changing the execution shell around the model, not the model itself. The code is on GitHub.
130
That's the last story.