Llama-3.1-70B-Instruct IQ4_XS

This file is exactly 37,902,666,304 bytes — 35.30 GiB / 37.90 GB at an effective 4.298 bits per weight. The nominal rate for IQ4_XS is lower; mixed-precision tensors make the real figure higher, always.

From the file· 1 file(s), summedFrom the file· 724 tensors parsed

Get it

35.30 GiB · 1 file
llama.cpp
llama-cli -hf bartowski/Meta-Llama-3.1-70B-Instruct-GGUF:IQ4_XS

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

Hugging Face CLI
hf download bartowski/Meta-Llama-3.1-70B-Instruct-GGUF Meta-Llama-3.1-70B-Instruct-IQ4_XS.gguf
Direct download

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
35.30 GiB
37.90 GB
Effective bpw
4.298
from real bytes ÷ params
Tensors
724
Header
7.87 MB
GGUF metadata

What this quantization actually contains

per-tensor types, parsed from the GGUF header
IQ4_XS
481
F32
162
Q5_K
80
Q6_K
1

A quantization label names a mixture, not a uniform precision. Attention and output tensors are routinely kept at higher precision than the label implies, which is exactly why the effective bits-per-weight above exceeds the nominal rate.

Llama-3.1-70B-Instruct IQ4_XS — exact size and VRAM — ossmodeldb