Model comparison
Qwen3.5-9B vs Mistral-Small-3.2-24B-Instruct-2506-ultra-uncensored-heretic
At Q4_K_M, Qwen3.5-9B is the smaller download — 5,627,044,256 bytes against 14,333,922,464. At long context the gap widens: Qwen3.5-9B's KV cache at 32K is 5.0× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| Qwen3.5-9B | Mistral-Small-3.2-24B-Instruct-2506-ultra-uncensored-heretic | |
|---|---|---|
| Parameters | 9.7B | 24.0B |
| Architecture | qwen35 | mistral3 |
| Layers | 32 | 40 |
| Native context | 262,144 | 131,072 |
| Mixture of experts | no | no |
| Quantizations published | 34 | 40 |
| Smallest quantization | 2.97 GiB | 4.91 GiB |
| Q4_K_M | 5.24 GiB | 13.35 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Qwen3.5-9B | Mistral-Small-3.2-24B-Instruct-2506-ultra-uncensored-heretic | Ratio |
|---|---|---|---|
| 4,096 | 0.13 GiB | 0.63 GiB | 5.00× |
| 8,192 | 0.25 GiB | 1.25 GiB | 5.00× |
| 16,384 | 0.50 GiB | 2.50 GiB | 5.00× |
| 32,768 | 1.00 GiB | 5.00 GiB | 5.00× |
| 65,536 | 2.00 GiB | 10.00 GiB | 5.00× |
| 131,072 | 4.00 GiB | 20.00 GiB | 5.00× |