Krea-2-Raw · gguf-org

Krea-2-Raw NVFP4

This file is exactly 10,436,821,888 bytes — 9.72 GiB / 10.44 GB at an effective 6.513 bits per weight. The nominal rate for NVFP4 is lower; mixed-precision tensors make the real figure higher, always.

From the file· 2 file(s), summed

Get it

9.72 GiB · 2 files

llama.cpp

llama-cli -hf gguf-org/krea-2-gguf:NVFP4

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

Hugging Face CLI

hf download gguf-org/krea-2-gguf --include "krea2_turbo-nvfp4.gguf"

This quantization is split across 2 files totalling 9.72 GiB. You need every part — the first is not a smaller version of the model.

Direct downloads — all 2 required

krea2_turbo-nvfp4.gguf
qwen3vl_4b-nvfp4.gguf

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

9.72 GiB

10.44 GB

Effective bpw

6.513

from real bytes ÷ params

Tensors

—

Header

—

GGUF metadata