WHAT NEW IN X11.04 WED

AI collections from X · 22 collections · 67,905 posts

Back to No.001 GPT-6 Astra in Practice: Tests, Use Cases, and FeedbackUpdated 9 hours ago
gpt-6IMAGE

Four models show 0.9 Spearman correlation on severity rankings

Testing shows severity ratings are highly correlated across GPT-6 Astra, GPT 5.6 Sol, Fable 5.1 and Gemini 3.8 Flash, with a Spearman coefficient of 0.9.

antra
@tessera_antraArchived 2026.09.19Originally posted 2026.09.19
Four models show 0.9 Spearman correlation on severity rankings
antra @tessera_antraOriginally posted 2026.09.19
Open original on X

Severity rankings are well-correlated between four judges, with 0.9 spearman correlation between GPT-6-Astra, Gpt 5.6 Sol, Fable 5.1 and Gemini 3.8 Flash.

x.com/tessera_antra/status/2101078654487019764Archived 2026.09.19 10:13
Open original on X