Model comparison

Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP vs Qwopus3.6-35B-A3B-v1

At Q4_K_M, Qwopus3.6-35B-A3B-v1 is the smaller download — 21,713,462,432 bytes against 36,545,830,912. At long context the gap widens: Qwopus3.6-35B-A3B-v1's KV cache at 32K is 3.2× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTPQwopus3.6-35B-A3B-v1
Parameters27.8B36.0B
Architectureqwen35clip
Layers6440
Native context262,144262,144
Mixture of expertsnoyes, 256 experts
Quantizations published3429
Smallest quantization10.12 GiB6.97 GiB
Q4_K_M34.04 GiB20.22 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTPQwopus3.6-35B-A3B-v1Ratio
4,0960.25 GiB0.08 GiB3.20×
8,1920.50 GiB0.16 GiB3.20×
16,3841.00 GiB0.31 GiB3.20×
32,7682.00 GiB0.63 GiB3.20×
65,5364.00 GiB1.25 GiB3.20×
131,0728.00 GiB2.50 GiB3.20×