Llama-3.1-8B-Instruct-heretic I1-IQ4_XS

This file is exactly 8,895,337,280 bytes — 8.28 GiB / 8.90 GB at an effective 8.862 bits per weight. The nominal rate for I1-IQ4_XS is lower; mixed-precision tensors make the real figure higher, always.

From the file· 2 file(s), summed

Get it

8.28 GiB · 2 files
llama.cpp
llama-cli -hf mradermacher/Llama-3.1-8B-Instruct-heretic-i1-GGUF:I1-IQ4_XS

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

Hugging Face CLI
hf download mradermacher/Llama-3.1-8B-Instruct-heretic-i1-GGUF --include "Llama-3.1-8B-Instruct-heretic.i1-IQ4_XS.gguf"

This quantization is split across 2 files totalling 8.28 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
8.28 GiB
8.90 GB
Effective bpw
8.862
from real bytes ÷ params
Tensors
Header
GGUF metadata
Llama-3.1-8B-Instruct-heretic I1-IQ4_XS — exact size and VRAM — ossmodeldb