Model comparison
Salience-1.5-Pro vs Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP
At Q4_K_M, Salience-1.5-Pro is the smaller download — 21,713,464,032 bytes against 36,545,830,912. At long context the gap widens: Salience-1.5-Pro'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
| Salience-1.5-Pro | Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP | |
|---|---|---|
| Parameters | 36.0B | 27.8B |
| Architecture | qwen35moe | qwen35 |
| Layers | 40 | 64 |
| Native context | 262,144 | 262,144 |
| Mixture of experts | yes, 256 experts | no |
| Quantizations published | 51 | 34 |
| Smallest quantization | 9.55 GiB | 10.12 GiB |
| Q4_K_M | 20.22 GiB | 34.04 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Salience-1.5-Pro | Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP | 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× |