Model comparison

GLM-Z1-9B-0414 vs Qwen3-8B

At Q4_K_M, Qwen3-8B is the smaller download — 5,027,783,488 bytes against 6,166,574,464. At long context the gap widens: GLM-Z1-9B-0414's KV cache at 32K is 3.6× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

GLM-Z1-9B-0414Qwen3-8B
Parameters9.4B8.2B
Architectureglm4qwen3
Layers4036
Native context32,76840,960
Mixture of expertsnono
Quantizations published4751
Smallest quantization2.98 GiB2.12 GiB
Q4_K_M5.74 GiB4.68 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
ContextGLM-Z1-9B-0414Qwen3-8BRatio
4,0960.16 GiB0.56 GiB3.60×
8,1920.31 GiB1.13 GiB3.60×
16,3840.63 GiB2.25 GiB3.60×
32,7681.25 GiB4.50 GiB3.60×
65,5362.50 GiB9.00 GiB3.60×
131,0725.00 GiB18.00 GiB3.60×