FLUX.2-klein-4B · unsloth

FLUX.2-klein-4B Q3_K_M

This file is exactly 2,124,489,280 bytes — 1.98 GiB / 2.12 GB at an effective 4.385 bits per weight. The nominal rate for Q3_K_M is lower; mixed-precision tensors make the real figure higher, always.

From the file· 1 file(s), summedFrom the file· 149 tensors parsed

Get it

1.98 GiB · 1 file
llama.cpp
llama-cli -hf unsloth/FLUX.2-klein-4B-GGUF:Q3_K_M

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

Hugging Face CLI
hf download unsloth/FLUX.2-klein-4B-GGUF flux-2-klein-4b-Q3_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
1.98 GiB
2.12 GB
Effective bpw
4.385
from real bytes ÷ params
Tensors
149
Header
0.01 MB
GGUF metadata

What this quantization actually contains

per-tensor types, parsed from the GGUF header
F32
60
Q3_K
50
Q4_K
30
BF16
9

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.