Aleksa Gordić (水平问题) @gordic_aleksa · 15 Aug 2026
Sequel for my H100 blog is out now. We implement Blackwell matmul for NVFP4 from scratch, and outperform cuBLAS by 4.7% at N=8192. This time we have something better than Hilbert curves - that exact-fits for the Nvidia hardware. Full blog post with all details: https://t.co/m931p1DUxj
119 319Views
626Likes
73Reposts
20Replies
9Quotes
0Bookmarks
Is that a lot?
4.96×vs this author's median24 056 views is typical
67Percentile for this authorof 6 recent posts
32.7×vs 10K–100K median3 644 views is typical
3.8× audienceReachviews ÷ followers
0.61%Engagement rateof viewers reacted