gpt-6POST
Joke post mocks model arms race and homemade benchmarks
A joke post contrasts Anthropic's Fable 5.1 and OpenAI's GPT-6 Astra with bored AI posters who respond with their own dubious benchmark.
klöss
@kloss_xyzArchived 2026.09.14Originally posted 2026.09.10
klöss @kloss_xyzOriginally posted 2026.09.10
Anthropic: meet Fable 5.1 (it's smart) OpenAI: meet GPT-6 Astra (it's smarter) bored AI bro: meet my ass benchmark
This post quotes another post: “Developer runs GPT-6 Astra in a loop for a week on a butt-modeling benchmark”
x.com/kloss_xyz/status/2097838995694919999Archived 2026.09.14 15:48