Model comparison

LTX-2.3 vs LTX-2

At Q4_K_M, LTX-2 is the smaller download — 12,840,118,752 bytes against 14,194,441,248.

From the file· summed bytes, KV per layer

Side by side

LTX-2.3LTX-2
Parameters9.2B18.9B
Architectureqwen35ltxv
Layers
Native context
Mixture of expertsnono
Quantizations published9228
Smallest quantization7.39 GiB7.48 GiB
Q4_K_M13.22 GiB11.96 GiB
Licence

KV cache by context

the term that decides long-context viability
ContextLTX-2.3LTX-2Ratio
4,096
8,192
16,384
32,768
65,536
131,072