Model comparison

Qwen3.8-27B vs gemma-4-26B-roleplay-v2-merged

At Q4_K_M, gemma-4-26B-roleplay-v2-merged is the smaller download — 16,796,015,616 bytes against 16,810,714,336.

From the file· summed bytes, KV per layer

Side by side

Qwen3.8-27Bgemma-4-26B-roleplay-v2-merged
Parameters27.8B25.2B
Architectureqwen35gemma4
Layers64
Native context262,144
Mixture of expertsnono
Quantizations published2234
Smallest quantization8.39 GiB7.72 GiB
Q4_K_M15.66 GiB15.64 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3.8-27Bgemma-4-26B-roleplay-v2-mergedRatio
4,0960.25 GiB
8,1920.50 GiB
16,3841.00 GiB
32,7682.00 GiB
65,5364.00 GiB
131,0728.00 GiB