Model comparison
gemma-7b vs Qwen3-VL-8B-Instruct-abliterated-v1
At Q4_K_M, Qwen3-VL-8B-Instruct-abliterated-v1 is the smaller download — 5,027,785,792 bytes against 5,329,758,592.
From the file· summed bytes, KV per layer
Side by side
| gemma-7b | Qwen3-VL-8B-Instruct-abliterated-v1 | |
|---|---|---|
| Parameters | 8.5B | 8.8B |
| Architecture | gemma | qwen3vl |
| Layers | 28 | — |
| Native context | 8,192 | — |
| Mixture of experts | no | no |
| Quantizations published | 31 | 48 |
| Smallest quantization | 2.01 GiB | 1.97 GiB |
| Q4_K_M | 4.96 GiB | 4.68 GiB |
| Licence | gemma | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | gemma-7b | Qwen3-VL-8B-Instruct-abliterated-v1 | Ratio |
|---|---|---|---|
| 4,096 | 1.75 GiB | — | — |
| 8,192 | 3.50 GiB | — | — |
| 16,384 | 7.00 GiB | — | — |
| 32,768 | 14.00 GiB | — | — |
| 65,536 | 28.00 GiB | — | — |
| 131,072 | 56.00 GiB | — | — |