Model comparison
Tiger-Gemma-9B-v1 vs Qwen3-8B
At Q4_K_M, Qwen3-8B is the smaller download — 5,027,783,488 bytes against 5,761,057,728. At long context the gap widens: Qwen3-8B's KV cache at 32K is 1.3× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| Tiger-Gemma-9B-v1 | Qwen3-8B | |
|---|---|---|
| Parameters | 9.2B | 8.2B |
| Architecture | gemma2 | qwen3 |
| Layers | 42 | 36 |
| Native context | 8,192 | 40,960 |
| Mixture of experts | no | no |
| Quantizations published | 20 | 51 |
| Smallest quantization | 3.20 GiB | 2.12 GiB |
| Q4_K_M | 5.37 GiB | 4.68 GiB |
| Licence | — | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Tiger-Gemma-9B-v1 | Qwen3-8B | Ratio |
|---|---|---|---|
| 4,096 | 1.31 GiB | 0.56 GiB | 2.33× |
| 8,192 | 2.05 GiB | 1.13 GiB | 1.82× |
| 16,384 | 3.36 GiB | 2.25 GiB | 1.49× |
| 32,768 | 5.99 GiB | 4.50 GiB | 1.33× |
| 65,536 | 11.24 GiB | 9.00 GiB | 1.25× |
| 131,072 | 21.74 GiB | 18.00 GiB | 1.21× |