WHAT NEW IN X10.29 THU

AI collections from X · 22 collections · 58,925 posts

gpt-6POST

Artificial Analysis asked to explain Astra's poor results, one user says

The poster demands that Artificial Analysis explain why GPT-6 Astra performed so poorly in its testing, which conflicts sharply with OpenAI's own benchmark results.

AshutoshShrivastava
@ai_for_successArchived 2026.09.14Originally posted 2026.09.04
AshutoshShrivastava @ai_for_successOriginally posted 2026.09.04
Open original on X

Artificial Analysis owes a very clear explanation of how GPT-6 / ASTRA performed so poorly in their testing while OpenAI had very different benchmark results. One of them has rigged the benchmarks result.

x.com/ai_for_success/status/2095732727660355665Archived 2026.09.14 15:48