Loading profile…

World & Business

Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

Sylvie L (@__sylvie_l__) · Digg

Sylvie L

@__sylvie_l__

make chips hot @ together

Featured in

AI

NyoomFloat12's lossless 12-bit weight compression reportedly speeds up LLM inference

A researcher reports 1.23× faster token generation on Qwen3-32B and up to 2.54× rollout throughput.

3101234567890123456789

7 Sources, 2h ago, first seen 2h ago

Three panels. (a) Per-bit entropy of BF16 weights in Llama-3.2-1B: exponent bits E7–E4 near zero, every other bit near 1. (b) The high bytes of two BF16 weights packed into one NF12 meta byte. (c) Rollout throughput against concurrency on Qwen3-32B: NF12 peaks at 787 tok/s, DFloat11 at 357, BF16 at 310.

Connections

@__sylvie_l__