Is GLM-5 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 1.0x smaller, because of how its attention layers cache.

Model comparison

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

the term that decides long-context viability

Context	GLM-5	GLM-5.2	Ratio
4,096	0.34 GiB	0.34 GiB	1.00×
8,192	0.69 GiB	0.69 GiB	1.00×
16,384	1.37 GiB	1.37 GiB	1.00×
32,768	2.74 GiB	2.74 GiB	1.00×
65,536	5.48 GiB	5.48 GiB	1.00×
131,072	10.97 GiB	10.97 GiB	1.00×