Model comparison

SCAIL-2 vs LTX-2.3

At Q4_K_M, SCAIL-2 is the smaller download — 10,862,812,832 bytes against 14,194,441,248.

From the file· summed bytes, KV per layer

Side by side

SCAIL-2LTX-2.3
Parameters16.4B9.2B
Architecturewanqwen35
Layers
Native context
Mixture of expertsnono
Quantizations published2692
Smallest quantization5.61 GiB7.39 GiB
Q4_K_M10.12 GiB13.22 GiB
Licencemit

KV cache by context

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