Is Laguna-M.1 or GLM-5.2 smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

GLM-5.2 — its KV cache at 32K context is 3.2x smaller, because of how its attention layers cache.

Model comparison

Laguna-M.1 vs GLM-5.2

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: GLM-5.2'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

	Laguna-M.1	GLM-5.2
Parameters	226B	753B
Architecture	laguna	glm-dsa
Layers	70	78
Native context	262,144	1,048,576
Mixture of experts	yes, 256 experts	yes, 256 experts
Quantizations published	2	20
Smallest quantization	223.63 GiB	169.33 GiB
Q4_K_M	—	—
Licence	apache-2.0	mit

KV cache by context

the term that decides long-context viability

Context	Laguna-M.1	GLM-5.2	Ratio
4,096	1.09 GiB	0.34 GiB	3.19×
8,192	2.19 GiB	0.69 GiB	3.19×
16,384	4.38 GiB	1.37 GiB	3.19×
32,768	8.75 GiB	2.74 GiB	3.19×
65,536	17.50 GiB	5.48 GiB	3.19×
131,072	35.00 GiB	10.97 GiB	3.19×