Model comparison
AuraFlow-v0.3 vs FLUX.2-klein-9B
At Q4_K_M, AuraFlow-v0.3 is the smaller download — 4,047,070,496 bytes against 5,027,783,648.
From the file· summed bytes, KV per layer
Side by side
| AuraFlow-v0.3 | FLUX.2-klein-9B | |
|---|---|---|
| Parameters | 6.9B | 9.1B |
| Architecture | aura | flux |
| Layers | — | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 14 | 29 |
| Smallest quantization | 2.23 GiB | 3.71 GiB |
| Q4_K_M | 3.77 GiB | 4.68 GiB |
| Licence | apache-2.0 | other |
KV cache by context
the term that decides long-context viability
| Context | AuraFlow-v0.3 | FLUX.2-klein-9B | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |