Black Forest Labs · 12B parameters · FLUX.1 [dev] Non-Commercial License
What GPU do you need to run FLUX.1 [dev]?Memory and cards.
FLUX.1 [dev] needs about 14 GB of GPU memory (VRAM) at its smallest (8-bit FP8), so a 16 GB card such as an RTX 4080 runs it. The cheapest card on our list that fits, an RTX 4060 Ti 16GB, costs about $0.14 an hour. At its largest (16-bit, full precision) it needs about 24 GB.
GPU memory
FLUX.1 [dev], by version.
Resolution, batch size, upscalers and extra LoRAs all add memory. How we work it out
| Version | Memory needed | Card size | Cards that fit | From |
|---|---|---|---|---|
| 8-bit FP8 | about 14 GB | 16–20 GB | RTX 4090, RTX 5090, RTX 3090, RTX 4080, RTX 3090 Ti, RTX 5080, RTX 5070 Ti, RTX 4070 Ti SUPER, RTX 5060 Ti 16GB, RTX 4060 Ti 16GB | $0.14/hr on an RTX 4060 Ti 16GB |
| 16-bit, full precision | about 24 GB | 24 GB | RTX 4090, RTX 5090, RTX 3090, RTX 3090 Ti | $0.155/hr on an RTX 3090 |
A top-quality image model. Its license limits commercial use of the model itself, so read it before using it in a business. The FLUX.1 [schnell] version is Apache 2.0.
Run it
Start FLUX.1 [dev] in minutes.
The rented machine downloads it straight from Hugging Face, not through your computer. On KernelMachine, Rent an Image maker with these files, below: they load into ComfyUI's model folders by themselves once it's ready. Already renting one? Pick it under Load from link on the rental's page, or run km load with each link. ComfyUI on a rented GPU
$ km load https://huggingface.co/Comfy-Org/flux1-dev/resolve/main/flux1-dev.safetensors
$ km load https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/t5xxl_fp16.safetensors
$ km load https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/clip_l.safetensors
$ km load https://huggingface.co/Comfy-Org/Lumina_Image_2.0_Repackaged/resolve/main/split_files/vae/ae.safetensors
Files to load
FLUX.1 [dev] in ComfyUI.
The files ComfyUI's own "Flux.1 Dev: Text to Image" template uses, 34.2 GB in all. Each goes in its own folder, and Load from link puts it there. Then open that template from ComfyUI's template menu.
| File | Folder | Size |
|---|---|---|
| flux1-dev.safetensors | models/diffusion_models | 23.8 GB |
| t5xxl_fp16.safetensors | models/text_encoders | 9.8 GB |
| clip_l.safetensors | models/text_encoders | 246 MB |
| ae.safetensors | models/vae | 335 MB |
They start loading as soon as the rental is ready, straight from Hugging Face to the machine. Already renting an Image maker? Pick FLUX.1 [dev] under Load from link on the rental's page. A card with 24 GB or more runs them.
Common questions
Questions, answered plainly.
How much VRAM does FLUX.1 [dev] need?
8-bit FP8: about 14 GB; 16-bit, full precision: about 24 GB.
Can an RTX 4090 run FLUX.1 [dev]?
Yes. FLUX.1 [dev] (16-bit, full precision) needs about 24 GB, which fits in the RTX 4090's 24 GB.
Can I run FLUX.1 [dev] on KernelMachine?
Yes. Rent an Image maker with these files, on this page, starts ComfyUI on the cheapest card with 16 GB or more, and the machine downloads each file from Hugging Face into ComfyUI's folders once it's ready. Already renting an Image maker? Pick FLUX.1 [dev] under Load from link on the rental's page.
Other models
Compare with other image and video models.
Preparing an invite-only beta
Get an email when access opens.
No launch date is set. We'll email you once when KernelMachine opens. Nothing else.