WHAT NEW IN X10.31 SAT

AI collections from X · 22 collections · 63,170 posts

Back to No.001 GPT-6 Astra in Practice: Tests, Use Cases, and FeedbackUpdated 5 hours ago
gpt-6VIDEO

Same prompt: GPT-6 Astra wins on realism, DeepSeek V4.1 Flash stuns on detail

Given the same prompt, GPT-6 Astra's output looked more realistic while DeepSeek V4.1 Flash, given 3–4 hours, produced impressively detailed results.

Salio
@Mr_SalioArchived 12 days agoOriginally posted 2026.10.17
Salio @Mr_SalioOriginally posted 2026.10.17
Open original on X

🚨 GPT-6 Astra vs DeepSeek V4.1 Flash Same prompt Two very impressive outputs Astra wins on realism, but DeepSeek absolutely COOKED with its own result I gave DeepSeek around 3-4 hours to work on this, and honestly, the wait was worth it The level of detail and the final result are seriously impressive This is a minor upgrade model competing with Astra and Opus 5.5 is insane

x.com/mr_salio/status/2111460511103565886Archived 2026.10.18 16:03
Open original on X