While upscaling, it aims to reduce unwanted blur and subtly enhance skin texture/details when needed. It maintains a high adherence to the original image.Currently a work in progress—feel free to test it out!
This workflow is based on my Flux checkpoint model Extra Realism Flux
Hyperrealistic Image Geneation Tool
This workflow creates an Isolated Image and then remove its background, save into PNG file.the output resolution by default is 3072 x 3072 upscaled by Ultimate SD.the RemBG model might not be perfect sometime, I would recommend using other Cut Out tool like Canva for some specific image.for normal generation please use this Lora : Isolated Object PNG Lora - v1.0-noShadow
# supported Flux models *GGUF all supported for better result not less than Q8 # supported Lora models All supported including custom loras DONT FORGET YOUR TRIGGER WORD at beginning of positive promptthe workflow is highly optimized for faster result processing and output the best result better than regular models 😉the enhanced comfyui custom config for even faster boosting for high end laptops and extra gpus or multi-gpu with npu is available can produce this results in few seconds availabel can contact me at brianlangay0@gmail.com ☕☕☕☕☕☕☕..
This workflow creates an Isolated Image.the output resolution by default is 3072 x 3072 upscaled by Ultimate SD.for normal generation please use this Lora : Isolated Object PNG Lora - v1.0-noShadow
HyDiT Based Image Geneation Tool
With this tool you can use your favorite SD model with flux finishingEnjoy it
SDXL Hyperrealistic Image Geneation Tool with Unique Look
Credits:Checkpoint: FLUX.1 - dev-fp8 | Stable Diffusion Model - Checkpoint | Tensor.ArtModel:Bianca: Bianca - Indonesian Influencer Girls - Flux - bianca-v1Elsa Japasal/Eca: Indonesian Woman - Elsa Japasal | Aura Echa - V8Dewi Persik: Dewi Perssik (LoRA) Indonesia - DePeAnya Geraldine: Anya geraldine - IndonesiaAriel Tatum: Indonesia - Tatum(fux)Anselma Putri: Anselma - Indonesian Influencer Girls - Flux - V1Freya JKT48: Freya JKT48 (LoRA) Indonesia - Flux v2Zee JKT48: Zee JKT48 - FluxLivy Renata: Indonesian Woman - Livy RenataMarion Jola: Indonesia - Jola(flux)Video Created using External AI Video Service, then used as AI Tools Cover, Not Generated with TensorArt.To create videos, first generate an image using this AI tools, then visit the website https://hailuoai.video/, create an account, and upload your image. Optionally, you can give a prompt to the video generator and then await the results.To use generated video as post or AI Tools cover you can run this workflow Upload Local Videos as Generation Result | ComfyUI WorkflowDisclaimer: In accordance with copyright laws, including but not limited to the Copyright Act of 1976 (U.S. Code Title 17) and similar legislation worldwide, this tool operates within the boundaries of "fair use" provisions. Fair use, as outlined in Section 107 of the U.S. Copyright Act, allows the limited use of copyrighted material for purposes such as commentary, criticism, research, parody, or educational activities.This tool creates transformative works for non-commercial purposes, adhering to fair use principles, and is not intended to infringe on the commercial market value of the original work or harm the reputation of the depicted individual. Commercial use of this tool or its outputs without proper licensing agreements may constitute copyright infringement.By using this tools, you agree to take full responsibility for any consequences arising from its use. I am not responsible for any misuse, infringement, or legal issues that may occur as a result of your usage.
this is just a test
Hyperrealistic Image Geneation Tool with Unique Look
#Now you can use multiple LoRAs with upscaling.Flux.1 is a state-of-the-art text-to-image model developed by Black Forest Labs. It's renowned for its ability to generate incredibly detailed and realistic images based on textual descriptions. The model is a significant advancement in the field of AI-powered image generation.Guidance scalesfor photography: 3.0 - 3.5for style; Around 2SAMPLER; EulerSCHEDULER; simple or beta
Credits:Checkpoint: FLUX.1 - dev-fp8 | Stable Diffusion Model - Checkpoint | Tensor.ArtModel: Flux realism Lora - v1 | Stable Diffusion Model - LoRA | Tensor.ArtVideo Created using External AI Video Service, then used as AI Tools Cover, Not Generated with TensorArt.To create videos, first generate an image using this AI tools, then visit the website https://hailuoai.video/, create an account, and upload your image. Optionally, you can give a prompt to the video generator and then await the results.To use generated video as post or AI Tools cover you can run this workflow Upload Local Videos as Generation Result | ComfyUI Workflow
Bring your precious memories back to life effortlessly with EraFix Lab - powered by TDNM! Simply upload your old photos, add a few details, and let our cutting-edge AI work its magic. No complex tools or skills needed—just a few clicks, and you'll get stunning, restored images that capture the essence of the past. Turn faded moments into vivid treasures in no time!*This is a premium app, you get 5 free uses to experience. To use more and enjoy a series of other premium apps of TDNM without limits, subscribe for only $5.9/month.----Hồi sinh những ký ức quý giá của bạn một cách dễ dàng với EraFix Lab - powered by TDNM! Chỉ cần tải lên những bức ảnh cũ, thêm một vài thông tin và để AI tiên tiến của chúng tôi thực hiện phần còn lại. Không cần đến các công cụ phức tạp hay kỹ năng chuyên môn—chỉ với vài thao tác đơn giản, bạn sẽ nhận được những bức ảnh phục dựng tuyệt đẹp, tái hiện trọn vẹn tinh thần của quá khứ. Biến những khoảnh khắc phai mờ thành những kỷ vật sống động trong chớp mắt!*Đây là ứng dụng cao cấp, bạn được sử dụng miễn phí 5 lần để trải nghiệm. Để sử dụng nhiều hơn và tận hưởng một loạt các ứng dụng cao cấp khác của TDNM mà không giới hạn, hãy đăng ký chỉ với $5,9/tháng.
Credits:Checkpoint: FLUX.1 - dev-fp8 | Stable Diffusion Model - Checkpoint | Tensor.ArtModel: Boreal-FD: (Boring Reality Flux-Dev LoRA) - 2.0 | Stable Diffusion Model - LoRA | Tensor.ArtThe Flux Boreal aims to replicate the look of raw phone-captured images, as opposed to the polished realism typical of the default Flux behavior.
Just drop a close up of a Face into the input and get a CHEAP - RANDOM - Halloween image back!!! Try a few and see what you like... Please post your results so I can see how they come out for you. The Output will be an Image and A Text Description of the prompt generated
Hyperrealistic Image Geneation Tool with Unique Look
***FIXED & UPDATED on September 26***Transform any image by simply describing what you want instead 😎The beauty of this img2img tool is that it detects the size of the original image and adjusts the output size automatically.Prompt:Unlike Stable Diffusion (SD), FLUX diffusion understands clear, natural language.For example, in the cover image of this tool, the prompt was simply:Mature woman sitting, surrounded by booksSampling Steps:FLUX is a powerful model that doesn't need as many steps as many SD checkpoints (that aren't Turbo or Lightening etc.):10 steps = good results15 steps = very good results20 steps = great results25 steps = can be excellent but sometimes image quality can start to degrade (presence of artifacts)>25 steps = not worth it IMOFor example, in the cover images of this tool, the number of steps was 10.Denoise (results may vary):0.1 to 0.5 = You won't see much change0.6 to 0.7 = Slight changes0.7 to 0.85 = Major changes>0.9 = Extreme changesThis tool's preset values:FLUX1 - dev-fp8max_shift 1.15base_shift 0.50euler betaCFG 1.0Note that there is no Negative prompt field, because FLUX doesn't use Negatives.Enjoy 😊>>> v1 of this tool here
Hyperrealistic Image Geneation Tool with Unique LookUpdate Notes :08102024 - Reworking the workflow, for more realistic tweaks-----------------------------------------------------------------------TIP :Just type your idea on the prompt box in natural language.
High Detailed images Generator
CNLeathermanDrawing suggestion: Adopt a new generation of FLUX1 dev schema, super clear: iteration step number 15-20, CFG fill in 3-6 according to your needs. The weight can be debugged by yourself. It is recommended to be between 0.4-1.1.Positive:1boy,mature man,LeatherBoi, leathersuit,muscle,LeatherBro
