Model comparison

Wan2.2-VACE-Fun-A14B vs LTX-2

At Q4_K_M, Wan2.2-VACE-Fun-A14B is the smaller download — 11,639,453,600 bytes against 12,840,118,752.

From the file· summed bytes, KV per layer

Side by side

Wan2.2-VACE-Fun-A14BLTX-2
Parameters17.3B18.9B
Architecturewanltxv
Layers
Native context
Mixture of expertsnono
Quantizations published2028
Smallest quantization7.31 GiB7.48 GiB
Q4_K_M10.84 GiB11.96 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextWan2.2-VACE-Fun-A14BLTX-2Ratio
4,096
8,192
16,384
32,768
65,536
131,072