Model comparison
Qwen3.8-27B vs magnum-v4-27b
At Q4_K_M, magnum-v4-27b is the smaller download — 16,645,381,728 bytes against 16,810,714,336. At long context the gap widens: Qwen3.8-27B's KV cache at 32K is 3.3× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| Qwen3.8-27B | magnum-v4-27b | |
|---|---|---|
| Parameters | 27.8B | 27.2B |
| Architecture | qwen35 | gemma2 |
| Layers | 64 | 46 |
| Native context | 262,144 | 8,192 |
| Mixture of experts | no | no |
| Quantizations published | 22 | 22 |
| Smallest quantization | 8.39 GiB | 7.82 GiB |
| Q4_K_M | 15.66 GiB | 15.50 GiB |
| Licence | apache-2.0 | gemma |
KV cache by context
the term that decides long-context viability
| Context | Qwen3.8-27B | magnum-v4-27b | Ratio |
|---|---|---|---|
| 4,096 | 0.25 GiB | 1.44 GiB | 5.75× |
| 8,192 | 0.50 GiB | 2.25 GiB | 4.49× |
| 16,384 | 1.00 GiB | 3.68 GiB | 3.68× |
| 32,768 | 2.00 GiB | 6.56 GiB | 3.28× |
| 65,536 | 4.00 GiB | 12.31 GiB | 3.08× |
| 131,072 | 8.00 GiB | 23.81 GiB | 2.98× |