Penulis

Brian

Staff Technical Writer

2 posts

Brian educates on using software and tools effectively. He is pragmatic in each of his articles and provides value in terms of work that can actually be done today.

$GGUF, GPTQ, AWQ, EXL2 quantization formats compared: how model weights, runtime overhead, and KV cache stack up in memory$

AI dan Machine Learning

GGUF, GPTQ, AWQ, EXL2: Bagaimana Format Kuantisasi LLM Sebenarnya Menggunakan Memori

Bandingkan penggunaan memori GGUF, GPTQ, AWQ, dan EXL2, dari ukuran file Q4_K_M hingga pertumbuhan KV cache dan overhead runtime.

Brian Jul 2, 2026 12 menit baca

Unified memory explained: discrete GPU memory requires a copy across PCIe between system RAM and VRAM, while unified memory is one shared pool the CPU and GPU both access directly

AI dan Machine Learning

Apa Itu Unified Memory, dan Mengapa Itu Memungkinkan Mini PC Menjalankan Model 235B?

Unified memory memungkinkan PC AI yang kompak memuat model kelas 235B yang tidak dapat ditampung oleh satu GPU 24-32GB pun. Apa itu, mengapa ini bekerja, dan mengapa lebih besar ti

Brian Jul 2, 2026 11 menit baca