Model comparison
FLUX.1-dev vs Chroma
At Q4_K_M, Chroma is the smaller download — 5,566,533,792 bytes against 6,931,817,760.
From the file· summed bytes, KV per layer
Side by side
| FLUX.1-dev | Chroma | |
|---|---|---|
| Parameters | 11.9B | 8.9B |
| Architecture | flux | flux |
| Layers | — | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 37 | 273 |
| Smallest quantization | 3.75 GiB | 3.99 GiB |
| Q4_K_M | 6.46 GiB | 5.18 GiB |
| Licence | other | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | FLUX.1-dev | Chroma | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |