Model comparison

polyglot-lion-0.6b vs MiniCPM5-1B-Claude-Opus-Fable5-Thinking

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

From the file· summed bytes, KV per layer

Side by side

polyglot-lion-0.6bMiniCPM5-1B-Claude-Opus-Fable5-Thinking
Parameters782M1.1B
Architectureqwen3vlllama
Layers24
Native context131,072
Mixture of expertsnono
Quantizations published2415
Smallest quantization0.19 GiB0.43 GiB
Q4_K_M0.64 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
Contextpolyglot-lion-0.6bMiniCPM5-1B-Claude-Opus-Fable5-ThinkingRatio
4,0960.09 GiB
8,1920.19 GiB
16,3840.38 GiB
32,7680.75 GiB
65,5361.50 GiB
131,0723.00 GiB