Model comparison
ZAYA1-8B-Coder vs gemma-4-12B-it-qat-q4_0-unquantized
These two publish different quantization sets; the table below has the exact sizes.
From the file· summed bytes, KV per layer
Side by side
| ZAYA1-8B-Coder | gemma-4-12B-it-qat-q4_0-unquantized | |
|---|---|---|
| Parameters | 8.8B | 12.0B |
| Architecture | zaya | gemma4 |
| Layers | 80 | 48 |
| Native context | 131,072 | 262,144 |
| Mixture of experts | yes, 16 experts | no |
| Quantizations published | 3 | 2 |
| Smallest quantization | 5.19 GiB | 6.50 GiB |
| Q4_K_M | 5.19 GiB | — |
| Licence | — | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | ZAYA1-8B-Coder | gemma-4-12B-it-qat-q4_0-unquantized | Ratio |
|---|---|---|---|
| 4,096 | 0.31 GiB | 0.72 GiB | 2.30× |
| 8,192 | 0.63 GiB | 0.97 GiB | 1.55× |
| 16,384 | 1.25 GiB | 1.47 GiB | 1.18× |
| 32,768 | 2.50 GiB | 2.47 GiB | 1.01× |
| 65,536 | 5.00 GiB | 4.47 GiB | 1.12× |
| 131,072 | 10.00 GiB | 8.47 GiB | 1.18× |