Model comparison
dolphin-2.9.1-mixtral-1x22b vs gemma-4-E4B-it
These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 13.8× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| dolphin-2.9.1-mixtral-1x22b | gemma-4-E4B-it | |
|---|---|---|
| Parameters | 22.2B | 8.0B |
| Architecture | llama | gemma4 |
| Layers | 56 | 42 |
| Native context | 65,536 | 131,072 |
| Mixture of experts | yes, 1 experts | no |
| Quantizations published | 45 | 36 |
| Smallest quantization | 4.49 GiB | 3.30 GiB |
| Q4_K_M | — | 4.64 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | dolphin-2.9.1-mixtral-1x22b | gemma-4-E4B-it | Ratio |
|---|---|---|---|
| 4,096 | 0.88 GiB | 0.12 GiB | 7.11× |
| 8,192 | 1.75 GiB | 0.18 GiB | 9.85× |
| 16,384 | 3.50 GiB | 0.29 GiB | 12.19× |
| 32,768 | 7.00 GiB | 0.51 GiB | 13.84× |
| 65,536 | 14.00 GiB | 0.94 GiB | 14.84× |
| 131,072 | 28.00 GiB | 1.82 GiB | 15.40× |