Model comparison
Le-Chaton-Slim-23B vs Qwen3.6-27B
At Q4_K_M, Le-Chaton-Slim-23B is the smaller download — 14,183,211,648 bytes against 16,547,397,888. At long context the gap widens: Qwen3.6-27B's KV cache at 32K is 1.6× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| Le-Chaton-Slim-23B | Qwen3.6-27B | |
|---|---|---|
| Parameters | 23.3B | 27.8B |
| Architecture | llama | qwen35 |
| Layers | 26 | 64 |
| Native context | 262,144 | 262,144 |
| Mixture of experts | yes, 10 experts | no |
| Quantizations published | 34 | 40 |
| Smallest quantization | 4.59 GiB | 8.74 GiB |
| Q4_K_M | 13.21 GiB | 15.41 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Le-Chaton-Slim-23B | Qwen3.6-27B | Ratio |
|---|---|---|---|
| 4,096 | 0.41 GiB | 0.25 GiB | 1.62× |
| 8,192 | 0.81 GiB | 0.50 GiB | 1.62× |
| 16,384 | 1.63 GiB | 1.00 GiB | 1.62× |
| 32,768 | 3.25 GiB | 2.00 GiB | 1.62× |
| 65,536 | 6.50 GiB | 4.00 GiB | 1.62× |
| 131,072 | 13.00 GiB | 8.00 GiB | 1.62× |