Model comparison
gemma-3n-E4B-it vs Qwythos-9B-Claude-Mythos-5-1M
At Q4_K_M, gemma-3n-E4B-it is the smaller download — 4,237,064,064 bytes against 11,516,776,704. At long context the gap widens: gemma-3n-E4B-it's KV cache at 32K is 2.0× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| gemma-3n-E4B-it | Qwythos-9B-Claude-Mythos-5-1M | |
|---|---|---|
| Parameters | 7.8B | 9.4B |
| Architecture | gemma3n | qwen35 |
| Layers | 35 | 32 |
| Native context | 32,768 | 1,048,576 |
| Mixture of experts | no | no |
| Quantizations published | 38 | 16 |
| Smallest quantization | 2.57 GiB | 5.38 GiB |
| Q4_K_M | 3.95 GiB | 10.73 GiB |
| Licence | gemma | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | gemma-3n-E4B-it | Qwythos-9B-Claude-Mythos-5-1M | Ratio |
|---|---|---|---|
| 4,096 | 0.11 GiB | 0.13 GiB | 1.14× |
| 8,192 | 0.16 GiB | 0.25 GiB | 1.52× |
| 16,384 | 0.27 GiB | 0.50 GiB | 1.83× |
| 32,768 | 0.49 GiB | 1.00 GiB | 2.03× |
| 65,536 | 0.93 GiB | 2.00 GiB | 2.15× |
| 131,072 | 1.80 GiB | 4.00 GiB | 2.22× |