Model comparison
Qwen3.5-4B-Claude-4.6-OS-Auto-Variable-HERETIC-UNCENSORED-THINKING vs Qwen3-VL-8B-Instruct-abliterated-v1
At Q4_K_M, Qwen3.5-4B-Claude-4.6-OS-Auto-Variable-HERETIC-UNCENSORED-THINKING is the smaller download — 2,707,515,520 bytes against 5,027,785,792.
From the file· summed bytes, KV per layer
Side by side
| Qwen3.5-4B-Claude-4.6-OS-Auto-Variable-HERETIC-UNCENSORED-THINKING | Qwen3-VL-8B-Instruct-abliterated-v1 | |
|---|---|---|
| Parameters | 4.5B | 8.8B |
| Architecture | qwen35 | qwen3vl |
| Layers | 32 | — |
| Native context | 262,144 | — |
| Mixture of experts | no | no |
| Quantizations published | 36 | 48 |
| Smallest quantization | 1.09 GiB | 1.97 GiB |
| Q4_K_M | 2.52 GiB | 4.68 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Qwen3.5-4B-Claude-4.6-OS-Auto-Variable-HERETIC-UNCENSORED-THINKING | Qwen3-VL-8B-Instruct-abliterated-v1 | Ratio |
|---|---|---|---|
| 4,096 | 0.13 GiB | — | — |
| 8,192 | 0.25 GiB | — | — |
| 16,384 | 0.50 GiB | — | — |
| 32,768 | 1.00 GiB | — | — |
| 65,536 | 2.00 GiB | — | — |
| 131,072 | 4.00 GiB | — | — |