Model comparison

grug-9b vs Qwythos-9B-Claude-Mythos-5-1M

At Q4_K_M, grug-9b is the smaller download — 5,629,109,056 bytes against 11,516,776,704.

From the file· summed bytes, KV per layer

Side by side

grug-9bQwythos-9B-Claude-Mythos-5-1M
Parameters9.4B9.4B
Architectureqwen35qwen35
Layers3232
Native context262,1441,048,576
Mixture of expertsnono
Quantizations published2816
Smallest quantization3.51 GiB5.38 GiB
Q4_K_M5.24 GiB10.73 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
Contextgrug-9bQwythos-9B-Claude-Mythos-5-1MRatio
4,0960.13 GiB0.13 GiB1.00×
8,1920.25 GiB0.25 GiB1.00×
16,3840.50 GiB0.50 GiB1.00×
32,7681.00 GiB1.00 GiB1.00×
65,5362.00 GiB2.00 GiB1.00×
131,0724.00 GiB4.00 GiB1.00×