Model comparison

GLM-OCR vs embeddinggemma-300m

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: embeddinggemma-300m's KV cache at 32K is 13.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

GLM-OCRembeddinggemma-300m
Parameters1.3B303M
Architectureglm4gemma-embedding
Layers1624
Native context131,0722,048
Mixture of expertsnono
Quantizations published3010
Smallest quantization0.34 GiB0.26 GiB
Q4_K_M0.51 GiB
Licence

KV cache by context

the term that decides long-context viability
ContextGLM-OCRembeddinggemma-300mRatio
4,0960.25 GiB0.04 GiB7.11×
8,1920.50 GiB0.05 GiB9.85×
16,3841.00 GiB0.08 GiB12.19×
32,7682.00 GiB0.14 GiB13.84×
65,5364.00 GiB0.27 GiB14.84×
131,0728.00 GiB0.52 GiB15.40×