Llama-3.1-70B-Instruct · MaziyarPanahi

Llama-3.1-70B-Instruct Q5_K_M

This file is exactly 49,949,816,960 bytes — 46.52 GiB / 49.95 GB at an effective 5.664 bits per weight. The nominal rate for Q5_K_M is lower; mixed-precision tensors make the real figure higher, always.

From the file· 1 file(s), summedFrom the file· 723 tensors parsed

Get it

46.52 GiB · 1 file
llama.cpp
llama-cli -hf MaziyarPanahi/Meta-Llama-3.1-70B-Instruct-GGUF:Q5_K_M

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

Hugging Face CLI
hf download MaziyarPanahi/Meta-Llama-3.1-70B-Instruct-GGUF Meta-Llama-3.1-70B-Instruct.Q5_K_M.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.

Repository →
Size
46.52 GiB
49.95 GB
Effective bpw
5.664
from real bytes ÷ params
Tensors
723
Header
7.86 MB
GGUF metadata

What this quantization actually contains

per-tensor types, parsed from the GGUF header
Q5_K
481
F32
161
Q6_K
81

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.