Cosmos-Predict2-2B-Text2Image · calcuis
Cosmos-Predict2-2B-Text2Image Q5_K_M
This file is exactly 19,516,964,448 bytes — 18.18 GiB / 19.52 GB at an effective 10.945 bits per weight. The nominal rate for Q5_K_M is lower; mixed-precision tensors make the real figure higher, always.
From the file· 8 file(s), summed
Get it
18.18 GiB · 8 files
llama.cpp
llama-cli -hf calcuis/cosmos-predict2-gguf:Q5_K_M
Downloads and runs in one step, resolving the quantization by name.
Hugging Face CLI
hf download calcuis/cosmos-predict2-gguf --include "cosmos-predict2-14b-t2i-ex3-q5_k_m.gguf"
This quantization is split across 8 files totalling 18.18 GiB. You need every part — the first is not a smaller version of the model.
Direct downloads — all 8 required
- cosmos-predict2-14b-t2i-ex3-q5_k_m.gguf
- cosmos-predict2-2b-t2i-ex3-q5_k_m.gguf
- cosmos-predict2-2b-t2i-ex-q5_k_m.gguf
- cosmos-predict2-2b-v2w-480p-10fps-ex3-q5_k_m.gguf
- cosmos-predict2-2b-v2w-480p-16fps-ex3-q5_k_m.gguf
- cosmos-predict2-2b-v2w-720p-10fps-ex3-q5_k_m.gguf
- cosmos-predict2-2b-v2w-720p-16fps-ex3-q5_k_m.gguf
- cosmos-predict2-2b-v2w-720p-16fps-ex-q5_k_m.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.
Size
18.18 GiB
19.52 GB
Effective bpw
10.945
from real bytes ÷ params
Tensors
—
Header
—
GGUF metadata