FLUX.1-Krea-dev vs stable-diffusion-v1-4
FLUX.1-Krea-dev (11.9B parameters) and stable-diffusion-v1-4 (860M parameters) side by side: the memory each needs at every precision, what it costs to run on a live GPU, and the context window, KV cache and license where they are published. Numbers are computed from the models' published specs; this page does not rank quality.
Side by side
| Fact | FLUX.1-Krea-dev | stable-diffusion-v1-4 |
|---|---|---|
| Parameters | 11.9B | 860M |
| Architecture | – | – |
| License | – | – |
| Published precision | BF16 | F32 |
| VRAM needed, As published | 26.6 GB | 3.8 GB |
| VRAM needed, FP8 | 13.3 GB | 1.0 GB |
| VRAM needed, INT4 | 6.7 GB | 0.5 GB |
| Cheapest live fit, As published | RTX 4080 Super · $0.338/hr | V100 · $0.088/hr |
| Cheapest live fit, FP8 | RTX 4000 SFF Ada · $0.198/hr | RTX 4070 Super · $0.121/hr |
| Cheapest live fit, INT4 | RTX 4070 Super · $0.121/hr | RTX 4070 Super · $0.121/hr |
VRAM is the weight size at each precision times a flat 1.2 overhead; see the methodology. The FP8 and INT4 rows need a quantized checkpoint or an engine that quantizes on load. The fit is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
Which to pick
- stable-diffusion-v1-4 needs less VRAM at its published precision (3.8 GB against 26.6 GB), so it fits on a smaller GPU.
- stable-diffusion-v1-4 has the cheaper live GPU fit at its published precision ($0.088/hr against $0.338/hr).
These follow only from the facts in the table above. Whether either model does your task well is a separate question this page does not answer.
Keep reading
- FLUX.1-Krea-dev: full VRAM table and live GPU fit
- stable-diffusion-v1-4: full VRAM table and live GPU fit
- The FLUX model series
- The Stable Diffusion model series
- All models that fit in 32 GB
- All models that fit in 8 GB
Other comparisons
