Griffith-7/nf4-triton-kernel

An optimized, high-performance NF4 (NormalFloat 4-bit) dequantization Triton GPU kernel achieving up to 1.41x speedup over bitsandbytes.

5 best points · 1 comments · 1 Hacker News post · 6 GitHub stars · View on GitHub

Hacker News discussions

• Show HN: Fast NF4 dequantization Triton kernel (1.41x faster than bitsandbytes) (5 points, 1 comments, by Griffith-7, 2026-07-15)