X · @teortaxesTex
· X / Twitter
Perhaps. Would be quite sad But really, doesn’t matter. They can make V4.5 with like 3 MLA layers. For now, they need to run their token factory hot,…
Perhaps. Would be quite sadBut really, doesn’t matter. They can make V4.5 with like 3 MLA layers. For now, they need to run their token factory hot, build strong models and elite datasets. When more compute comes online, they can experiment with richer architectures.GDP: DeepSeek V4-Flash even after recent very impress