Model comparison

InternVL3_5-14B vs Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP

At Q4_K_M, InternVL3_5-14B is the smaller download — 9,001,749,952 bytes against 36,545,830,912.

From the file· summed bytes, KV per layer

Side by side

InternVL3_5-14BQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP
Parameters15.1B27.8B
Architectureqwen3qwen35
Layers64
Native context262,144
Mixture of expertsnono
Quantizations published2834
Smallest quantization4.62 GiB10.12 GiB
Q4_K_M8.38 GiB34.04 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextInternVL3_5-14BQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTPRatio
4,0960.25 GiB
8,1920.50 GiB
16,3841.00 GiB
32,7682.00 GiB
65,5364.00 GiB
131,0728.00 GiB