Model comparison
Serenity-31B-v1.1 vs Qwen-AgentWorld-35B-A3B
These two publish different quantization sets; the table below has the exact sizes.
From the file· summed bytes, KV per layer
Side by side
| Serenity-31B-v1.1 | Qwen-AgentWorld-35B-A3B | |
|---|---|---|
| Parameters | 32.7B | 34.7B |
| Architecture | gemma4 | qwen35moe |
| Layers | — | 40 |
| Native context | — | 262,144 |
| Mixture of experts | no | yes, 256 experts |
| Quantizations published | 6 | 15 |
| Smallest quantization | 14.24 GiB | 10.71 GiB |
| Q4_K_M | 17.40 GiB | — |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Serenity-31B-v1.1 | Qwen-AgentWorld-35B-A3B | Ratio |
|---|---|---|---|
| 4,096 | — | 0.08 GiB | — |
| 8,192 | — | 0.16 GiB | — |
| 16,384 | — | 0.31 GiB | — |
| 32,768 | — | 0.63 GiB | — |
| 65,536 | — | 1.25 GiB | — |
| 131,072 | — | 2.50 GiB | — |