MiniMax-H3 · unsloth

MiniMax-H3 Q4_K

This file is exactly 22,801,760,448 bytes — 21.24 GiB / 22.80 GB at an effective 5.507 bits per weight. The nominal rate for Q4_K is lower; mixed-precision tensors make the real figure higher, always.

From the file· 2 file(s), summed

Get it

21.24 GiB · 2 files
llama.cpp
llama-cli -hf unsloth/MiniMax-H3-GGUF:Q4_K

Downloads and runs in one step, resolving the quantization by name.

Hugging Face CLI
hf download unsloth/MiniMax-H3-GGUF --include "minimax_h3_fl2va_pruned-Q4_K.gguf"

This quantization is split across 2 files totalling 21.24 GiB. You need every part — the first is not a smaller version of the model.

Direct downloads — all 2 required

Straight from the Hugging Face CDN — we host nothing and earn nothing from this. Verify what you received against the exact byte count above; a size mismatch is the usual cause of a file that will not load.

Size
21.24 GiB
22.80 GB
Effective bpw
5.507
from real bytes ÷ params
Tensors
Header
GGUF metadata
MiniMax-H3 Q4_K — exact size and VRAM — ossmodeldb