Model comparison
North-Mini-Code-1.0 vs gemma-4-E4B-it
At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 18,744,024,640. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 2.2× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| North-Mini-Code-1.0 | gemma-4-E4B-it | |
|---|---|---|
| Parameters | 30.5B | 8.0B |
| Architecture | cohere2moe | gemma4 |
| Layers | 49 | 42 |
| Native context | 500,000 | 131,072 |
| Mixture of experts | yes, 128 experts | no |
| Quantizations published | 41 | 36 |
| Smallest quantization | 7.93 GiB | 3.30 GiB |
| Q4_K_M | 17.46 GiB | 4.64 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | North-Mini-Code-1.0 | gemma-4-E4B-it | Ratio |
|---|---|---|---|
| 4,096 | 0.38 GiB | 0.12 GiB | 3.11× |
| 8,192 | 0.52 GiB | 0.18 GiB | 2.92× |
| 16,384 | 0.72 GiB | 0.29 GiB | 2.52× |
| 32,768 | 1.13 GiB | 0.51 GiB | 2.23× |
| 65,536 | 1.94 GiB | 0.94 GiB | 2.06× |
| 131,072 | 3.57 GiB | 1.82 GiB | 1.96× |