Model comparison

Qwen3.8-27B vs Qwopus3.6-27B-v2-MTP-GGUF

At Q4_K_M, Qwopus3.6-27B-v2-MTP-GGUF is the smaller download — 9,374,507,456 bytes against 16,810,714,336.

From the file· summed bytes, KV per layer

Side by side

Qwen3.8-27BQwopus3.6-27B-v2-MTP-GGUF
Parameters27.8B27.3B
Architectureqwen35qwen35
Layers64
Native context262,144
Mixture of expertsnono
Quantizations published224
Smallest quantization8.39 GiB8.73 GiB
Q4_K_M15.66 GiB8.73 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3.8-27BQwopus3.6-27B-v2-MTP-GGUFRatio
4,0960.25 GiB
8,1920.50 GiB
16,3841.00 GiB
32,7682.00 GiB
65,5364.00 GiB
131,0728.00 GiB