Michal Piszczek

mentions 2 type Person feed RSS

// recent coverage 2 mentions

14:29

2026-08-17

piszczek.pl

large-language-models

Qwen3.8-27B at 256K on a 24GB RTX PRO 4000 SFF (432 GB/s): 50 tok/s with MTP

A custom llama.cpp build with Qwen3.8-27B's embedded MTP drafter achieves 50.44 tokens per second on an NVIDIA RTX PRO 4000 Blackwell SFF with 24 GB VRAM, a 21.97% gain over clean master (55.40 vs 45.…

16:40

2026-08-01

dev.to

ai-tools

How Much Does AI Actually Cost? The Field Guide to 12 AI Economics Calculators

Developer Michal Piszczek released a suite of 12 free AI economics calculators at piszczek.pl/tools, designed to quantify AI costs, energy use, and agent verification. The tools include a token cost c…

// co-occurs with top 8 entities

Qwen3.8-27B 1 NVIDIA RTX PRO 4000 Blackwell SFF 1 llama.cpp 1 RTX 2000 Ada 1 CUDA 12.9.86 1 GCC 14.2 1 piszczek.pl 1 Joule Wars 1

// topics top 6 topics

ai infrastructure 2 large language models 1 ai research 1 ai tools 1 ai agents 1 developer tools 1