Model comparison
KAT-Coder-V2.5-Dev vs Qwen3.6-27B
At Q4_K_M, Qwen3.6-27B is the smaller download — 16,547,397,888 bytes against 21,391,448,480. At long context the gap widens: KAT-Coder-V2.5-Dev's KV cache at 32K is 3.2× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| KAT-Coder-V2.5-Dev | Qwen3.6-27B | |
|---|---|---|
| Parameters | 34.7B | 27.8B |
| Architecture | qwen35moe | qwen35 |
| Layers | 40 | 64 |
| Native context | 262,144 | 262,144 |
| Mixture of experts | yes, 256 experts | no |
| Quantizations published | 29 | 40 |
| Smallest quantization | 9.11 GiB | 8.74 GiB |
| Q4_K_M | 19.92 GiB | 15.41 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | KAT-Coder-V2.5-Dev | Qwen3.6-27B | Ratio |
|---|---|---|---|
| 4,096 | 0.08 GiB | 0.25 GiB | 3.20× |
| 8,192 | 0.16 GiB | 0.50 GiB | 3.20× |
| 16,384 | 0.31 GiB | 1.00 GiB | 3.20× |
| 32,768 | 0.63 GiB | 2.00 GiB | 3.20× |
| 65,536 | 1.25 GiB | 4.00 GiB | 3.20× |
| 131,072 | 2.50 GiB | 8.00 GiB | 3.20× |