Mistral-Large-3-675B-Instruct-2512 · bartowski

Mistral-Large-3-675B-Instruct-2512 IQ3_XS

This file is exactly 279,130,750,432 bytes — 259.96 GiB / 279.13 GB at an effective 3.316 bits per weight. The nominal rate for IQ3_XS is lower; mixed-precision tensors make the real figure higher, always.

From the file· 8 file(s), summed

Get it

259.96 GiB · 8 files
llama.cpp
llama-cli -hf bartowski/mistralai_Mistral-Large-3-675B-Instruct-2512-GGUF:IQ3_XS

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

Hugging Face CLI
hf download bartowski/mistralai_Mistral-Large-3-675B-Instruct-2512-GGUF --include "mistralai_Mistral-Large-3-675B-Instruct-2512-IQ3_XS/mistralai_Mistral-Large-3-675B-Instruct-2512-IQ3_XS-*-of-*.gguf"

This quantization is split across 8 files totalling 259.96 GiB. You need every part — the first is not a smaller version of the model.

Direct downloads — all 8 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.

Repository →
Size
259.96 GiB
279.13 GB
Effective bpw
3.316
from real bytes ÷ params
Tensors
—
Header
—
GGUF metadata
Mistral-Large-3-675B-Instruct-2512 IQ3_XS — exact size and VRAM — ossmodeldb