Flux 1.1 Pro

Flux is an open-source AI image generation model created by Black Forest Labs, the team behind Stable Diffusion. It produces highly photorealistic images from text prompts and is available in several variants, including FLUX 1.1 Pro (paid API access), Flux.1 Dev (open-weight for developers), and Flux.1 Schnell (fastest, free for personal use). Black Forest Labs has since shipped the FLUX.2 line, and FLUX 3 is what the vendor describes as its latest model family, while the FLUX 1.1 Pro documentation now sits under Legacy Models. Flux competes directly with Midjourney and DALL-E 3 on image quality.

You can access Flux through several platforms. For the easiest experience, use Black Forest Labs' own browser playground at dashboard.bfl.ai, or a third-party web interface such as Replicate, fal.ai, or Freepik's Flux integration, just type a prompt and generate. Developers can run Flux.1 Dev locally using ComfyUI or the diffusers Python library with a GPU. Flux.1 Pro is available via API through Black Forest Labs and partner platforms.

Flux.1 Schnell is completely free for personal use, you can run it locally or use it on free-tier platforms. Flux.1 Dev is open-weight and free to download but has a non-commercial license. FLUX 1.1 Pro requires paid API access and costs $0.04 per image on Black Forest Labs' own pay-as-you-go API, with no subscription and no seat fee. Many platforms offer a small number of free Flux generations daily.

Pricing depends on which Flux variant and platform you use. FLUX 1.1 Pro costs $0.04 per image on Black Forest Labs' own pay-as-you-go API, and FLUX 1.1 Pro Ultra costs $0.06 per image. Flux.1 Dev is free to run locally if you have a compatible GPU (12GB+ VRAM recommended). Platform pricing varies, Freepik includes Flux in its subscription, while Replicate and fal.ai charge per-generation. Running locally has no per-image cost beyond your hardware and electricity.

Midjourney produces more stylized, artistic images and has a simpler Discord-based workflow. Flux excels at photorealism and offers more flexibility, you can run it locally, fine-tune it, and integrate it into custom workflows via API. Midjourney costs $10-60/month with no free tier. FLUX has open-weights releases you can run locally for free, and hosted API pricing is pay as you go. Choose Midjourney for consistent artistic style, Flux for photorealism and developer control.

Flux was created by Black Forest Labs, a company founded by the original creators of Stable Diffusion (Robin Rombach, Andreas Blattmann, and others from the CompVis lab at LMU Munich). They left Stability AI to build Flux as a next-generation image model. Black Forest Labs is headquartered in Freiburg, Germany, and has raised significant funding from Andreessen Horowitz.

Flux supports flexible resolutions. Standard outputs include 1024×1024, 1024×768, and 768×1024. Flux.1 Pro can generate images up to 2048×2048 resolution. When running locally, output size is limited by your GPU VRAM. For best results, use aspect ratios close to 1:1, 4:3, or 3:4, extreme aspect ratios may reduce quality.

It depends on the variant. Flux.1 Pro includes full commercial rights for generated images. Flux.1 Schnell (Apache 2.0 license) allows commercial use. Flux.1 Dev has a non-commercial license, you can use it for research and personal projects but not for commercial products without a separate license from Black Forest Labs.

Flux produces more photorealistic results and gives you more control over outputs, especially when running locally with ComfyUI. DALL-E 3 is easier to use (built into ChatGPT) and handles text rendering in images better. DALL-E 3 also has built-in safety guardrails. Flux is more flexible and cost-effective for high-volume generation, while DALL-E 3 is more convenient for casual users.

Yes, if you have a compatible GPU. Flux.1 Dev requires about 12GB of VRAM minimum (an NVIDIA RTX 3060 or higher). Flux.1 Schnell can run on 8GB VRAM with quantized models. Install ComfyUI or use the Hugging Face diffusers library in Python. Running locally gives you unlimited free generations, full privacy, and the ability to fine-tune the model on your own data.