Model comparison
Qwen3-VL-4B-Instruct-c_abliterated-v2 vs Qwen3-4B
At Q4_K_M, Qwen3-4B is the smaller download — 2,497,280,256 bytes against 2,497,282,848.
From the file· summed bytes, KV per layer
Side by side
| Qwen3-VL-4B-Instruct-c_abliterated-v2 | Qwen3-4B | |
|---|---|---|
| Parameters | 4.4B | 4.0B |
| Architecture | qwen3vl | qwen3 |
| Layers | — | 36 |
| Native context | — | 40,960 |
| Mixture of experts | no | no |
| Quantizations published | 12 | 29 |
| Smallest quantization | 1.55 GiB | 1.01 GiB |
| Q4_K_M | 2.33 GiB | 2.33 GiB |
| Licence | apache-2.0 | — |
KV cache by context
the term that decides long-context viability
| Context | Qwen3-VL-4B-Instruct-c_abliterated-v2 | Qwen3-4B | Ratio |
|---|---|---|---|
| 4,096 | — | 0.56 GiB | — |
| 8,192 | — | 1.13 GiB | — |
| 16,384 | — | 2.25 GiB | — |
| 32,768 | — | 4.50 GiB | — |
| 65,536 | — | 9.00 GiB | — |
| 131,072 | — | 18.00 GiB | — |