\n \n\n# Run FLUX.1 [dev] on Novita AI\n**GitHub List: Novita AI Templates Catalogue\n\n## What is FLUX.1 [dev]?\n\nThe FLUX.1 [dev] model is a 12 billion parameter text-to-image model that uses a rectified flow transformer architecture to generate images from text descriptions. The platform requires users to agree to a non-commercial license in order to access its functionalities.\n\n \n\n## Key Features\n\n- Cutting-edge output quality, second only to FLUX.1 [pro].\n \n- Competitive prompt following, matching closed-source alternatives.\n \n- Trained using guidance distillation, making FLUX.1 [dev] more efficient.\n \n- Open weights to drive research and enable innovative workflows.\n \n- Usable for personal, scientific, and commercial purposes under the FLUX.1 [dev] Non-Commercial License.\n \n- Available via API from multiple providers, and integrated into ComfyUI and Diffusers.\n \n- Limitations:\n \n - Not intended for factual information, may amplify biases.\n \n - Restrictions on use, including prohibitions on harmful activities.\n \n\n## What is the Intended Use of FLUX.1 [dev]?\n\nDevelopers and artists aim to generate unique visual content from text prompts. It can be used in graphic design, advertising, content creation, and research.\n\n## Technical Details of FLUX.1 [dev]\n\n### Model Architecture\n\nFLUX.1 [dev] uses a 12 billion parameter rectified flow transformer architecture capable of generating images from text descriptions.\n\n### Training Data\n\nFLUX.1 [dev] utilizes diverse data sets from multiple sources, including public image libraries and edited collections.\n\n### Performance Benchmarks\n\nFLUX.1 [dev] demonstrates excellent performance in:\n\n- Adhering to prompts\n \n- Visual quality\n \n- Anatomical accuracy\n \n- Handling complex scenes\n \n- Supporting multiple aspect ratios and resolutions from 0.1 to 2.0 megapixels.\n \n\n## Comparing FLUX.1 [dev] with Other Models\n\n### Accuracy and Quality\n\n- FLUX.1 [dev]: Offers solid performance and image quality, slightly below the Pro version but superior to mainstream SD 3 Ultra models.\n \n- FLUX.1 [pro]: The most powerful model, delivering the highest quality in image generation, making it a choice for demanding applications.\n \n- FLUX.1 [schnell]: The lightest and fastest version, optimized for speed and efficiency, suitable for applications where quick processing is prioritized over maximum image\n \n\n### Pricing\n\n- FLUX.1 [dev]: $0.030/image\n \n- FLUX.1 [pro]: $0.055/image\n \n- FLUX.1 [schnell]: $0.003/image\n \n\n### Targeted Users\n\n- FLUX.1 [dev]: Commonly used for development and testing. Ideal for users who need high-quality output but do not require the absolute highest performance.\n \n- FLUX.1 [pro]: Designed for professional users who demand the highest detail and quality of images.\n \n- FLUX.1 [schnell]: Optimized for users with limited resources.\n \n\n## API Endpoint\n\nThe FLUX.1 [dev] is available on Novita AI. We offer:\n\n - FLUX.1 \[dev\] - Text to Image\n \n- FLUX.1 \[dev\] - Image to Image\n \n- FLUX.1 \[dev\] LoRA - Text to Image\n \n- FLUX.1 \[dev\] Realism LoRA - Text to Image\n \n- FLUX.1 \[schnell\] - Text to Image\n \n- FLUX.1 \[schnell\] - Image to Image\n\n## Run on Novita AI\n\nFLUX.1 [dev] is now available on the Novita AI Instance! Get started quickly without any installation—setup takes just a minute or two. Enjoy a scalable solution capable of running models efficiently and affordably. Try out our Novita AI template today!\n\n## How Can You Use FLUX.1 [dev] with the Diffusers Python Library?\n\n \n\nTo use FLUX.1 [dev] with the diffusers python library, first install or upgrade diffusers\n bash\\npip install -U diffusers\\n\nThen you can use FluxPipeline to run the model\n\npython\\nimport torch\\nfrom diffusers import FluxPipeline\\npipe = FluxPipeline.from_pretrained(\"black-forest-labs/FLUX.1-dev\", torch_dtype=torch.bfloat16)\\npipe.enable_model_cpu_offload() #save some VRAM by offloading the model to CPU. Remove this if you have enough GPU power\\nprompt = \"A cat holding a sign that says hello world\"\\nimage = pipe(\\n prompt,\\n height=1024,\\n width=1024,\\n guidance_scale=3.5,\\n num_inference_steps=50,\\n max_sequence_length=512,\\n generator=torch.Generator(\"cpu\").manual_seed(0)\\n).images[0]\\nimage.save(\"flux-dev.png\")\\n\nTo learn more check out the diffusers documentation.\n\n----------\n\n \n\n## What Does a Typical FLUX.1 [dev] Workflow Look Like?\n\n plain text\\n {\"last_node_id\":36,\"last_link_id\":57,\"nodes\":[{\"id\":33,\"type\":\"CLIPTextEncode\",\"pos\":[390,400],\"size\":{\"0\":422.84503173828125,\"1\":164.31304931640625},\"flags\":{\"collapsed\":true},\"order\":4,\"mode\":0,\"inputs\":[{\"name\":\"clip\",\"type\":\"CLIP\",\"link\":54,\"slot_index\":0,\"label\":\"clip\"}],\"outputs\":[{\"name\":\"CONDITIONING\",\"type\":\"CONDITIONING\",\"links\":[55],\"slot_index\":0,\"label\":\"CONDITIONING\"}],\"title\":\"CLIP Text Encode (Negative Prompt)\",\"properties\":{\"Node name for S&R\":\"CLIPTextEncode\"},\"widgets_values\":[\"\"],\"color\":\"#322\",\"bgcolor\":\"#533\"},{\"id\":27,\"type\":\"EmptySD3LatentImage\",\"pos\":[471,455],\"size\":{\"0\":315,\"1\":106},\"flags\":{},\"order\":0,\"mode\":0,\"outputs\":[{\"name\":\"LATENT\",\"type\":\"LATENT\",\"links\":[51],\"shape\":3,\"slot_index\":0,\"label\":\"LATENT\"}],\"properties\":{\"Node name for S&R\":\"EmptySD3LatentImage\"},\"widgets_values\":[1024,1024,1],\"color\":\"#323\",\"bgcolor\":\"#535\"},{\"id\":35,\"type\":\"FluxGuidance\",\"pos\":[576,96],\"size\":{\"0\":211.60000610351562,\"1\":58},\"flags\":{},\"order\":5,\"mode\":0,\"inputs\":[{\"name\":\"conditioning\",\"type\":\"CONDITIONING\",\"link\":56,\"label\":\"conditioning\"}],\"outputs\":[{\"name\":\"CONDITIONING\",\"type\":\"CONDITIONING\",\"links\":[57],\"shape\":3,\"slot_index\":0,\"label\":\"CONDITIONING\"}],\"properties\":{\"Node name for S&R\":\"FluxGuidance\"},\"widgets_values\":[3.5]},{\"id\":8,\"type\":\"VAEDecode\",\"pos\":[1151,195],\"size\":{\"0\":210,\"1\":46},\"flags\":{},\"order\":7,\"mode\":0,\"inputs\":[{\"name\":\"samples\",\"type\":\"LATENT\",\"link\":52,\"label\":\"samples\"},{\"name\":\"vae\",\"type\":\"VAE\",\"link\":46,\"label\":\"vae\"}],\"outputs\":[{\"name\":\"IMAGE\",\"type\":\"IMAGE\",\"links\":[9],\"slot_index\":0,\"label\":\"IMAGE\"}],\"properties\":{\"Node name for S&R\":\"VAEDecode\"}},{\"id\":34,\"type\":\"Note\",\"pos\":[831,501],\"size\":{\"0\":282.8617858886719,\"1\":164.08004760742188},\"flags\":{},\"order\":1,\"mode\":0,\"properties\":{\"text\":\"\"},\"widgets_values\":[\"Note that Flux dev and schnell do not have any negative prompt so CFG should be set to 1.0. Setting CFG to 1.0 means the negative prompt is ignored.\"],\"color\":\"#432\",\"bgcolor\":\"#653\"},{\"id\":30,\"type\":\"CheckpointLoaderSimple\",\"pos\":[48,192],\"size\":{\"0\":315,\"1\":98},\"flags\":{},\"order\":2,\"mode\":0,\"outputs\":[{\"name\":\"MODEL\",\"type\":\"MODEL\",\"links\":[47],\"shape\":3,\"slot_index\":0,\"label\":\"MODEL\"},{\"name\":\"CLIP\",\"type\":\"CLIP\",\"links\":[45,54],\"shape\":3,\"slot_index\":1,\"label\":\"CLIP\"},{\"name\":\"VAE\",\"type\":\"VAE\",\"links\":[46],\"shape\":3,\"slot_index\":2,\"label\":\"VAE\"}],\"properties\":{\"Node name for S&R\":\"CheckpointLoaderSimple\"},\"widgets_values\":[\"flux1-dev-fp8.safetensors\"]},{\"id\":9,\"type\":\"SaveImage\",\"pos\":[1375,194],\"size\":{\"0\":985.3012084960938,\"1\":1060.3828125},\"flags\":{},\"order\":8,\"mode\":0,\"inputs\":[{\"name\":\"images\",\"type\":\"IMAGE\",\"link\":9,\"label\":\"images\"}],\"properties\":{},\"widgets_values\":[\"ComfyUI\"]},{\"id\":6,\"type\":\"CLIPTextEncode\",\"pos\":[384,192],\"size\":{\"0\":422.84503173828125,\"1\":164.31304931640625},\"flags\":{},\"order\":3,\"mode\":0,\"inputs\":[{\"name\":\"clip\",\"type\":\"CLIP\",\"link\":45,\"label\":\"clip\"}],\"outputs\":[{\"name\":\"CONDITIONING\",\"type\":\"CONDITIONING\",\"links\":[56],\"slot_index\":0,\"label\":\"CONDITIONING\"}],\"title\":\"CLIP Text Encode (Positive Prompt)\",\"properties\":{\"Node name for S&R\":\"CLIPTextEncode\"},\"widgets_values\":[\"cute anime girl with massive fluffy fennec ears and a big fluffy tail blonde messy long hair blue eyes wearing a maid outfit with a long black gold leaf pattern dress and a white apron mouth open placing a fancy black forest cake with candles on top of a dinner table of an old dark Victorian mansion lit by candlelight with a bright window to the foggy forest and very expensive stuff everywhere there are paintings on the walls\"],\"color\":\"#232\",\"bgcolor\":\"#353\"},{\"id\":31,\"type\":\"KSampler\",\"pos\":[816,192],\"size\":{\"0\":315,\"1\":262},\"flags\":{},\"order\":6,\"mode\":0,\"inputs\":[{\"name\":\"model\",\"type\":\"MODEL\",\"link\":47,\"label\":\"model\"},{\"name\":\"positive\",\"type\":\"CONDITIONING\",\"link\":57,\"label\":\"positive\"},{\"name\":\"negative\",\"type\":\"CONDITIONING\",\"link\":55,\"label\":\"negative\"},{\"name\":\"latent_image\",\"type\":\"LATENT\",\"link\":51,\"label\":\"latent_image\"}],\"outputs\":[{\"name\":\"LATENT\",\"type\":\"LATENT\",\"links\":[52],\"shape\":3,\"slot_index\":0,\"label\":\"LATENT\"}],\"properties\":{\"Node name for S&R\":\"KSampler\"},\"widgets_values\":[972054013131368,\"randomize\",20,1,\"euler\",\"simple\",1]}],\"links\":[[9,8,0,9,0,\"IMAGE\"],[45,30,1,6,0,\"CLIP\"],[46,30,2,8,1,\"VAE\"],[47,30,0,31,0,\"MODEL\"],[51,27,0,31,3,\"LATENT\"],[52,31,0,8,0,\"LATENT\"],[54,30,1,33,0,\"CLIP\"],[55,33,0,31,2,\"CONDITIONING\"],[56,6,0,35,0,\"CONDITIONING\"],[57,35,0,31,1,\"CONDITIONING\"]],\"groups\":[],\"config\":{},\"extra\":{\"ds\":{\"scale\":1,\"offset\":[56.42885371989581,-14.294664184783073]}},\"version\":0.4}\\n \n## Frequently Asked Questions\n\n### What are the differences between FLUX.1 [pro] and FLUX.1 [dev]?\n\nFLUX.1 [pro] is a high-quality image generation model suited for professional applications, while FLUX.1 [dev] is optimized for efficiency, ideal for development and experimentation.\n\n \n\n### How does FLUX.1 [dev] handle various text descriptions for image generation?\n\nFLUX.1 [dev] processes text descriptions by encoding them, understanding context through attention mechanisms, and generating high-quality images that match the input. It adapts to various prompts effectively.\n\n### What applications can benefit from using FLUX.1 [dev]?\n\nFLUX.1 [dev] can enhance applications in art, marketing, game design, content creation, e-commerce, education, VR/AR, and film, generating high-quality images from text prompts.\n\n \n\n### What is Flux software development?\n\nFlux is a declarative coordination language that simplifies concurrency, eliminating the need for threads or locks, allowing quick integration of C/C++ components.\n\n \n\n### Is Flux better than Redux?\n\nWhether Flux is better than Redux depends on your needs. Redux offers simpler architecture and an easier learning curve, while Flux provides more flexibility.\n\n \n\n# License\n\nThis model falls under the FLUX.1 \[dev\] Non-Commercial License.\n\nGet in Touch:**\n\n- Email: iris@novita.ai\n \n- Discord: novita.ai\n---\n\n> Novita AI is the All-in-one cloud platform that empowers your AI ambitions. Integrated APIs, serverless, GPU Instance — the cost-effective tools you need. Eliminate infrastructure, start free, and make your AI vision a reality.