Model comparison

privacy-filter-multilingual vs Qwen3-8B

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

From the file· summed bytes, KV per layer

Side by side

privacy-filter-multilingualQwen3-8B
Parameters1.4B8.2B
Architectureopenai-privacy-filterqwen3
Layers836
Native context131,07240,960
Mixture of expertsyes, 128 expertsno
Quantizations published251
Smallest quantization2.62 GiB2.12 GiB
Q4_K_M4.68 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextprivacy-filter-multilingualQwen3-8BRatio
4,0960.56 GiB
8,1921.13 GiB
16,3842.25 GiB
32,7684.50 GiB
65,5369.00 GiB
131,07218.00 GiB