Model comparison
gemma-2b vs Qwen3-VL-8B-Instruct-abliterated-v1
At Q4_K_M, gemma-2b is the smaller download — 1,495,245,728 bytes against 5,027,785,792.
From the file· summed bytes, KV per layer
Side by side
| gemma-2b | Qwen3-VL-8B-Instruct-abliterated-v1 | |
|---|---|---|
| Parameters | 2.5B | 8.8B |
| Architecture | gemma | qwen3vl |
| Layers | 18 | — |
| Native context | 8,192 | — |
| Mixture of experts | no | no |
| Quantizations published | 38 | 48 |
| Smallest quantization | 0.84 GiB | 1.97 GiB |
| Q4_K_M | 1.39 GiB | 4.68 GiB |
| Licence | cc-by-nc-4.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | gemma-2b | Qwen3-VL-8B-Instruct-abliterated-v1 | Ratio |
|---|---|---|---|
| 4,096 | 0.07 GiB | — | — |
| 8,192 | 0.14 GiB | — | — |
| 16,384 | 0.28 GiB | — | — |
| 32,768 | 0.56 GiB | — | — |
| 65,536 | 1.13 GiB | — | — |
| 131,072 | 2.25 GiB | — | — |