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-0.6b | MiniCPM5-1B-Claude-Opus-Fable5-Thinking | |
|---|---|---|
| Parameters | 782M | 1.1B |
| Architecture | qwen3vl | llama |
| Layers | — | 24 |
| Native context | — | 131,072 |
| Mixture of experts | no | no |
| Quantizations published | 24 | 15 |
| Smallest quantization | 0.19 GiB | 0.43 GiB |
| Q4_K_M | — | 0.64 GiB |
| Licence | mit | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | polyglot-lion-0.6b | MiniCPM5-1B-Claude-Opus-Fable5-Thinking | Ratio |
|---|---|---|---|
| 4,096 | — | 0.09 GiB | — |
| 8,192 | — | 0.19 GiB | — |
| 16,384 | — | 0.38 GiB | — |
| 32,768 | — | 0.75 GiB | — |
| 65,536 | — | 1.50 GiB | — |
| 131,072 | — | 3.00 GiB | — |