Model comparison
Qwen3.6-27B vs GLM-Z1-32B-0414-uncensored-heretic-v2
At Q4_K_M, Qwen3.6-27B is the smaller download — 16,547,397,888 bytes against 19,686,117,248.
From the file· summed bytes, KV per layer
Side by side
| Qwen3.6-27B | GLM-Z1-32B-0414-uncensored-heretic-v2 | |
|---|---|---|
| Parameters | 27.8B | 32.6B |
| Architecture | qwen35 | glm4 |
| Layers | 64 | 61 |
| Native context | 262,144 | 32,768 |
| Mixture of experts | no | no |
| Quantizations published | 40 | 9 |
| Smallest quantization | 8.74 GiB | 14.80 GiB |
| Q4_K_M | 15.41 GiB | 18.33 GiB |
| Licence | apache-2.0 | mit |
KV cache by context
the term that decides long-context viability
| Context | Qwen3.6-27B | GLM-Z1-32B-0414-uncensored-heretic-v2 | Ratio |
|---|---|---|---|
| 4,096 | 0.25 GiB | 0.24 GiB | 1.05× |
| 8,192 | 0.50 GiB | 0.48 GiB | 1.05× |
| 16,384 | 1.00 GiB | 0.95 GiB | 1.05× |
| 32,768 | 2.00 GiB | 1.91 GiB | 1.05× |
| 65,536 | 4.00 GiB | 3.81 GiB | 1.05× |
| 131,072 | 8.00 GiB | 7.63 GiB | 1.05× |