FLUX.1 is an open-source AI image generation model launched by Black Forest Labs, a laboratory founded by Robin Rombach, a former core member of Stability AI. Many team members also come from the original Stable Diffusion development team. FLUX.1 has set a new benchmark for AI text-to-image models with its exceptional image generation capabilities. It offers three different variant versions to meet diverse user needs. With a training parameter count of up to 12 billion, it far surpasses the 2 billion of SD3 Medium and outperforms other popular models like Midjourney and DALL-E 3 across various key metrics.
FLUX.1 Models
To ensure that FLUX.1 also runs on computers with limited graphics card memory, there are a large number of differently detailed models. Black Forrest Labs offers three basic model variants:
FLUX.1 [schnell]
The [schnell] fast model is available for users with limited graphics card memory or little patience. Images can be generated in 4 steps in the shortest possible time with little VRAM. The price is lower image quality and depth of detail. The generated images may be used commercially (Apache 2.0 licence).
FLUX.1 [dev]
The [dev] model for non-commercial image generation. You can download the model as an open source checkpoint under the FLUX.1 non-commercial licence from Huggingface and install it on your own computer. The quality of the images generated is comparable to the results of the [pro] model.
FLUX.1 [pro]
The [pro] model of Flux AI allows professional, commercial applications with the highest image quality and depth of detail. The model can be used for a fee via the Flux API or via cloud services such as Replicate, freepic, fal.ai or togehter.ai.
FLUX.1 Kontext [dev]
The [pro] model of Flux AI allows professional, commercial applications with the highest image quality and depth of detail. The model can be used for a fee via the FLUX.1 API or via cloud services such as Replicate, freepic, fal.ai or togehter.ai.
Flux AI Model-Variations and Encoder Cheat Sheet
Dev Safetensors
| Precision | File Size | Steps | VRAM Req' | Source | Download |
|---|---|---|---|---|---|
FP16 | 24 GB | 20 - 30 | 24 GB | Black Forrest Labs | Download |
FP8 | 17 GB | 20 - 30 | 12 GB | lllyasviel | Download |
NF4 | 12 GB | 20 - 30 | 6 GB | lllyasviel | Download |
Dev Encoder
| Precision | File Size | Steps. | VRAM Req' | Source | Download |
|---|---|---|---|---|---|
VAE | 0.3 GB | Black Forrest Labs | Download | ||
T5 FP16 | 10 GB | > 32 GB RAM | lllyasviel | Download | |
T5 FP8 | 5 GB | < 32 GB RAM | lllyasviel | Download | |
Clip_l | 0.25 GB | lllyasviel | Download |
Schnell Safetensors
| Precision | File Size | Steps | VRAM Req' | Source | Download |
|---|---|---|---|---|---|
FP16 | 24 GB | 1 - 4 | 24 GB | Black Forrest Labs | Download |
FP8 | 12 GB | 1 - 4 | 12 GB | Kijai | Download |
NF4 | 12 GB | 1 - 4 | 6 GB | silveroxides | Download |