4.8 is meaningfully sharper than 4.6, not just on STEM
6:29 PM · Jun 3, 2026 · 5K Views
DeepSeek 4.8 Model Delivers Sharper Performance Than 4.6 Beyond STEM
--0--
Sentiment
Sentiment building, check back later.
Cluster Engagement
-
Views
-
Comments
-
Reposts
-
Bookmarks
Expand data
Posts from X
Most Activity
Most ActivityTimeline
VIEWS2.9KBOOKMARKS2REPLIES1
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
4.8 is meaningfully sharper than 4.6, not just on STEM
9hViews 2.9KLikes 11Bookmarks 2
LIKES17
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
It reminds me of Kimi-K2 negativity is strangely correlated with active vocabulary and conceptual boldness but boy does it get ANNOYING stop trying to DUNK ON ME BITCH
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
6hViews 1.7KLikes 17Bookmarks 1