WHAT NEW IN X10.30 FRI

AI collections from X · 22 collections · 61,986 posts

Back to No.001 GPT-6 Astra in Practice: Tests, Use Cases, and FeedbackUpdated 4 hours ago
gpt-6IMAGE

Astra's 99.9% vs 62.7% ARC-AGI-3 scores come down to scaffolding

The same model on the same test scored 99.9% and 62.7% depending on the scaffolding around it, and ARC Prize says it is not claiming AGI.

TNW
@thenextwebArchived 2026.09.14Originally posted 2026.09.08
Astra's 99.9% vs 62.7% ARC-AGI-3 scores come down to scaffolding
TNW @thenextwebOriginally posted 2026.09.08
Open original on X

OpenAI published 99.9%. The benchmark's own software returned 62.7%. Same model, same test, different scaffolding around it. ARC Prize printed both numbers and says it is not claiming AGI. https://thenextweb.com/news/openai-astra-arc-agi-3-harness-62-7-vs-99-9-benchmark-revisions

x.com/thenextweb/status/2097074513485680716Archived 2026.09.14 15:56
Open original on X