Comfyuiproiecte

Listen up, English Fluent, AI-enhanced IDE loving, automation pros. We’re a fast-moving startup building our entire operation on n8n, comfyui, civitai, open art workflows, social media platforms, and a handful of other puzzle pieces. • Digital Content Focus: We'll begin your job with building out content production workflows. • Part-time and Project based to start: We’ll keep your plate full refining and building new comfyui, n8n & open art workflows, tweaking API integrations, and making sure our systems are as automated as possible. • Growth on the horizon: We’re small and our budget is tight, but as we scale, so does this position—likely morphing into a full-time gig sooner than you’d expect. • Fully remote: ...

I already have a working ComfyUI-based image-generation setup, but it needs to run reliably at scale. My priority is to harden the existing pipeline so it can handle large batch jobs, quick regenerations, and seamless switching between several pre-trained models without crashing or slowing to a crawl. Here’s what I need from you: • Diagnose and eliminate the crashes and bottlenecks that appear during high-volume or repeat runs. • Optimise GPU/CPU memory usage so multiple models can be loaded, swapped, or unloaded cleanly. • Script or extend ComfyUI workflows for fully automated batch generation and one-click regeneration, tying in any helper utilities you find useful (Python, Bash, Docker, etc.). • Document the final setup clearly enough that ...

...a captivating, AI-generated music video that tells a visual story from start to finish. The mood is purely entertainment, so the images should feel cinematic and immersive rather than promotional or instructional. You’ll take the audio track I provide, develop a narrative concept around it, and use the generative tools you’re most comfortable with—Runway, Stable Diffusion, Midjourney, Pika, ComfyUI, After Effects, Premiere Pro, or any combination that gives the smoothest result. I’m open to mixed techniques (2D, 3D, or stylized footage) as long as the storyline stays clear and the visuals remain in sync with the beat and lyrical cues. Once the visuals are generated, I’ll need you to handle the full edit: sequencing, transitions, colour work, su...

...a captivating, AI-generated music video that tells a visual story from start to finish. The mood is purely entertainment, so the images should feel cinematic and immersive rather than promotional or instructional. You’ll take the audio track I provide, develop a narrative concept around it, and use the generative tools you’re most comfortable with—Runway, Stable Diffusion, Midjourney, Pika, ComfyUI, After Effects, Premiere Pro, or any combination that gives the smoothest result. I’m open to mixed techniques (2D, 3D, or stylized footage) as long as the storyline stays clear and the visuals remain in sync with the beat and lyrical cues. Once the visuals are generated, I’ll need you to handle the full edit: sequencing, transitions, colour work, su...

Animate an image (that i own) using ComfyUI (an ai app/website). Warning: the image that i want to animate is a soft nsfw since there's a young boy touching a girl's butt.

I have a project that should work with ComfyUI / WAN2 set-up and now need to turn it into approximately 15–20 minutes of finished, classroom-ready video. We have text scripts and the narration, on-screen text and any AI voices must be delivered in four Indian languages: Hindi, Tamil, Kannada and Malayalam. Your task is to take my existing workflow, refine or extend it where necessary and generate the complete video sequence—from asset creation right through to final render—so I can publish immediately. Fast turnaround is essential; I want to move as soon as the files are quality-checked and approved. Deliverables • 15-20 minutes of HD video in the four target languages (separate language cuts are fine) • Source ComfyUI workflow files and an...

...work fastest with). 2. Ping Razorpay’s APIs to confirm payment. 3. Forward the image to a ComfyUI workflow running on a RunPod serverless endpoint that leverages OOTDiffusion or ControlNet for the actual virtual try-on. 4. Return the final dressed-model image to the same WhatsApp conversation. Key expectations • You’ve set up ComfyUI on RunPod before and can show me a working endpoint. • You’re comfortable with at least one of the above WhatsApp Business APIs and can demonstrate prior message-automation work. • You can share a portfolio link or repo containing AI image manipulation samples—ideally virtual try-ons. Deliverables • Fully deployed serverless endpoint on RunPod with ComfyUI, pre-loaded with my...

High-threshold, high-requirement bounty for experienced AI model developers. Reward: Starting from $2500 USD (negotiable based on quality and experience), private discussion for details. I need a professional-grade checkpoint model (XL/Pony/Pony V7/Illustrious base acceptable), realistic style with full NSFW support, quality and style matching or surpassing , fully compatible with ComfyUI. Single-image generation in basic txt2img workflow ≤10s on RTX 5090 (32GB VRAM). Model size preferably SDXL level (6-8GB). Key focuses: Excellent limb/anatomy stability (no deformities, natural proportions) High prompt adherence (gender, ethnicity, body type, poses, expressions, scenes, etc.) Strong NSFW pose/scenario support with high consistency InstantID/ECOMID compatibility for fac...

Quiero automatizar...es contar con un flujo en ComfyUI que, al recibir esa foto, detecte la prenda, la recorte correctamente y la inserte en una composición flat lay con un fondo limpio y balanceado, replicando el estilo de la referencia que adjunto. El resultado final debe ser un JPG o PNG listo para subirse a mi catálogo de productos, manteniendo la fidelidad de los colores y las texturas originales. Necesito el archivo JSON del workflow de ComfyUI y una breve guía de uso para que pueda reaplicarlo con otras imágenes de la misma prenda sin mayor ajuste. Si durante el proceso se requieren nodos o extensiones específicas, indícalos en la guía para poder instalarlos. Acepto sugerencias de optimización siempre que el fl...

...and short videos—with a realistic, everyday-life aesthetic. Your role is to craft and refine the prompts that make the most of today’s diffusion and generative video models, control the tone so it remains natural rather than over-stylised, and document exactly how each shot or clip can be reproduced on demand. Feel free to use any tools for Stable Diffusion (A1111 + ControlNet + LoRA support), ComfyUI, and Runway/MagicAnimate for video. If you prefer Midjourney, DALL·E, Pika Labs, or similar platforms for specific shots then feel free to use so. The core of the assignment is simple: supply images and videos packs that reliably showcase photo-realistic NSFW scenes based on the characteristics that we decide and they should be grounded in everyday scenarios&mda...

I will provide Photos and Videos with different quality and format of a certain person. All data provided should be organized (useable for training or not) and processed (extract frames from Videos, if needed) to train an AI model for use with Stable Diffusion (ComfyUI) to create Photos and Videos of that person. The content created through the AI model should be: - As realistic as possible - As consistent as possible to create the Character trained with - Create high quality content - Uncensored All Data provided and created is confidential. It is prohibited to share any data, use it in any different way than fulfilling the purpose described above and has to be completely deleted as soon as the project is finished.

...high-quality word-art (script + decorative elements) on Etsy and other platforms. Samples can be viewed at this URL: I need a qualified ComfyUI developer to build a workflow that reliably generates sellable word-art images from text phrases, probably using Flux Schnell (or another model that would work well) with good spelling and consistent style. Environment & Constraints 2. OS / Hardware • Windows 11 • GPU: NVIDIA RTX 3060 12GB • Plenty of disk space 3. Existing ComfyUI install • Main install folder: C:Art Geyser0 ComfyUI Engine • Uses a virtualenv at C:Art Geyser0 ComfyUI Enginevenv • ComfyUI-Manager is already present and used. 4. Target Models & Node Packs Base model. Through my research it seems that...

Need Expert to Remotely Set Up ComfyUI + Stable Video Infinity on RunPod Description: I need an experienced ComfyUI / RunPod / AI video freelancer who can help me set up everything remotely. Here’s what I need: 1. Setup ComfyUI on my RunPod instance Make sure all folders, dependencies, and GPU settings are correct Fix any errors and make it run smoothly 2. Install Wan2.1 I2V 14B model + SVI models Place the downloaded Wan-AI/Wan2.1-I2V-14B-480P model in the correct folder Install Stable Video Infinity (SVI) models Connect them to the correct ComfyUI workflow 3. Setup working SVI workflow (image-to-video + long video) Import the official SVI workflow Test it to make sure it loads without

Summary I am looking for an experienced developer (AI, automation, ComfyUI) to finalize and professionalize my current devotional video automation system. I already have a functioning local n8n workflow that generates devotional videos via local comfyui wan2.2 (text + audio + 10 image prompts), but now I need three major upgrades: TASK 1 — Automatic Multi-Platform Upload System Goal: Fully automate upload of generated videos to: Facebook Multiple Instagram accounts TikTok WhatsApp Business LinkedIn Threads YouTube Requirements: Accepts video files from an existing local workflow (folder or webhook input). API-based upload, reliable, with proper handling of metadata (caption, tags, cover image). If possible: Scheduling option (e.g., “post daily at 09:00”). ...

...I need your help setting up ComfyUI on my Windows laptop so I can generate long, consistent AI videos (10 minutes total) for a project. My PC Specs: -GPU: NVIDIA RTX 4070 Laptop GPU (8GB VRAM) -CPU: Intel i9-14900HX -RAM: 32GB -Windows 11 Video Style I Need: I’m creating a 10-minute Roblox-style POV adventure video. The POV shows the same character with emerald-green tech gloves sliding through a giant neon fun-slide park. I need every 10-second generation to match perfectly in: -hands style -POV angle -slide theme -color palette -lighting -character proportions -world design Basically, the video should look like one continuous scene, even though it’s generated in small segments. What I need from you: Please build and install a complete ComfyUI...

I'm looking for an experienced ComfyUI developer to create a complete /ComfyUI workflow to create images similar to the ones attached I'm using a RTX 3060 GPU with 12 GB of ram, so it must be capable of operating within those parameters. would be my preferred model but that''s not entirely necessary if you have another that would work with my GPU. Prompt accuracy and spelling is of paramount that end my advisor proposes this as a suggested workflow. Stabilize typography by generating a clean text layout first (SVG/PNG) and then using that as a strong conditioning signal via ControlNet / IP-Adapter in ComfyUI, rather than asking the diffusion model to “invent” letters from scratch. Run a two-stage pipeline: 1) low-res layout for co...

I'm looking for an experienced ComfyUI developer to create a complete /ComfyUI workflow to create images similar to the ones attached I'm using a RTX 3060 GPU with 12 GB of ram, so it must be capable of operating within those parameters. would be my preferred model but that''s not entirely necessary if you have another that would work with my GPU. Prompt accuracy and spelling is of paramount importance. We will give it a list of phrases and it will generate a batch of images based upon that list. All nodes must be provided either by a valid download link or with the finalized product. All bids must be final any placeholder low bids will be rejected (no exceptions). If you want to bid on this and have the experience, please begin your bid with the p...

I need to modify the video_wan2_2_14B_s2v to extend its duration from the default 14 seconds to 60 seconds. The additional time should be used to extend existing scenes. Ideal Skills and Experience: - Experience in video editing - Proficiency with ComfyUI or similar tools - Attention to detail and creativity in scene extension

I have Stable Diffusion installed and have been using Automaticiii for a couple of months. I tried installing Comfyui myself(with help from Chatgpt) and had it running for a few days, but messed with settings and could not get in

...images and video (patience and creativity are crucial) Storyboarding skills to visually plan scenes and narratives An interest in filmmaking, especially visual storytelling, shot composition, lighting, and color grading Digital file organization and asset management Comfort with on-screen acting, motion reference, or facial performance capture Bonus Skills Hands-on experience with tools like ComfyUI Animation or VFX experience (not required, but welcome) We’re Looking for Someone Who Is... Creative and visually perceptive A clear communicator Organized and self-motivated Passionate about storytelling, generative media, and cinematic arts To Apply ***Firstly, tell us why you want to work in the field of AI-generated movies/series *** Also, please send us: A lin...

Title: Custom SDXL LoRA Character + Full Cloud Setup (RunPod/ComfyUI) Description: I’m looking for an experienced AI developer to create a high-quality custom LoRA character (SDXL preferred) using 20–40 reference images. The model will be used commercially (Fanvue/Patreon), so strong face and body consistency is essential. I currently use RunPod/, so the entire solution must run in the cloud. I need everything fully installed, configured, tested, and ready to use with zero technical setup required on my end. A short onboarding session (call or recorded tutorial) is required so I can operate the workflow independently. The workflow should also be prepared so I can later move it to a local workstation. Scope of work: Train a realistic SDXL LoRA with high identity stabi...

...optimized for realistic indoor and lifestyle visuals. OBJECTIVE : Build a workflow that: Extends any input image to a target aspect ratio (16:9, 21:7, 4:1, etc.). Preserves subject proportions, lighting, and environment continuity. Fills new canvas areas with realistic, seamless background generation. Supports automated or semi-automated batch processing via Python or ComfyUI. EXPECTED DELIVERABLES : Working AI workflow (ComfyUI graph or Python script). Documentation for setup, dependencies, and usage. At least 3 visual test results demonstrating clean, extended outputs in different aspect ratios. The images must fit the frame and expand the open areas. for example while converting from square to landscape the core image needs to cover the centre. (Optional) Si...

...Face-Swap / Identity-Transfer Expert (PhotoMaker V2 / InstantID / ComfyUI) Description: I need an experienced AI-image engineer or diffusion artist to perform realistic face-swaps in existing photographs. The final images must look completely natural — same lighting, background, clothes, and body — only the face changes. Deliverables: • Swap the face in target photos with the provided reference face. • Preserve lighting, shadows, reflections, and color tone exactly as in the original. • Ensure the final result looks like a real photo, not AI-generated. • Deliver high-resolution JPEG or PNG files (plus layered PSD or workflow file if possible). Requirements: • Experience with PhotoMaker V2, InstantID, IP-Adapter Face, ComfyUI, or...

Hello. This is my project: Every 15 minutes, an automated system will generate a new 5–10 minute video. • ChatGPT will create the prompt, YouTube title, description, and SEO tags. • ComfyUI will generate the video, audio, and thumbnail, then upload everything automatically to YouTube • The videos will rotate between nature sounds, white noise, study music, binaural tones, ASMR, ambience rooms, pet sounds, and more • The whole pipeline must run on a instance and be easy to replicate for future channels

I’m looking to set up a smart, end-to-end ComfyUI workflow that can turn a single prompt into ready-to-print visual posters in English, French, and Arabic. The pipeline should automatically handle text placement, font switching for RTL language support, and generate high-quality images sized for both online sharing and large-format printing. I am not a graphic designer, so the workflow should handle all design tasks automatically — text positioning, color selection, layout balancing, and font styling — producing a complete, ready-to-use image. It will mainly be used for educational and training visuals, but it should be flexible enough to adapt to general products or topics (phones, caps, cups, etc.). The workflow should: Automatically detect and arrange text (ti...

...recommended integration into the ComfyUI workflow, such as via LLM Node plugin or API call) Image Description Generation: Based on product information and selling points, LLM outputs positive prompts (e.g., "bright, vibrant, summer beach") and negative prompts (e.g., "dim, blurry, cluttered") Copywriting Generation: Generates promotional copy tailored to the target audience based on selling points (must specify tone, such as lively/professional, word limit, etc.) 2. Image Generation Module Tool Requirements: Based on the ComfyUI node-based workflow, must be compatible with the following functions: Reading and decoding cloud/local material libraries (supports common formats: images, PDFs, Word documents) Inputting the generated prompts into the ...

I already have a functional ComfyUI pipeline, but I need a more robust, high-throughput workflow that locks animated characters—both humanoid and non-humanoid—onto live-action facial and body skeletons. Right now results are hit-and-miss and I’d like a cleaner, repeatable solution. Here’s what I’m aiming to achieve next: Character-to-skeleton accuracy – Fine-tune or redesign my existing Controlnet setup so each frame faithfully tracks facial expressions and full-body movement. – Provide clear node graphs and parameter notes so I can reuse the template on large shot batches. Consistent environments (optional scope) – Guidance on generating or importing matching backdrops with AI tools, Blender, or Maya. Character placeme...

I need a comfyUI workflow, please check the attached document for detailed specifications

Job Title: Need Developer to Edit Existing ComfyUI Workflow on RunPod Job Description: I’m looking for someone experienced with ComfyUI and RunPod to make edits to an already existing workflow. The workflow is mostly set up, but I need some adjustments and improvements made to fit my current project requirements. What I need: Review and understand my existing ComfyUI workflow. Implement the requested edits or optimizations. Ensure everything runs smoothly within RunPod. Provide a brief explanation of what was changed and why. Requirements: Solid experience with ComfyUI workflows. Familiarity with RunPod setup and environment. Ability to debug or optimize nodes and workflow connections. Good communication and quick turnaround time. Nice to Hav...

...generated speech. • Safety & control – Ensure every high-privilege action (file edits, network calls, installs) is echoed back to the terminal before execution, giving me a clear veto option. – Provide a toggled “admin mode” so the AI runs elevated only when explicitly authorised. • Toolchain The project should continue to rely on free or open-source components: Tortoise, XTTS, Piper, ComfyUI, OpenSeeFace, plus any additional Python packages you deem essential. Deliverables 1. A Git repo with clean, commented code and setup scripts for Windows. 2. Step-by-step installation notes (pip requirements, environment variables, model downloads). 3. Short screen-capture demo showing voice command → AI reasoning → self-co...

...scene). Text generation / adaptation: Optionally use local LLMs (e.g. Wan 2.2‑based text models) to create or adapt the devotional texts in German/English. Integrate a TTS service (e.g. ElevenLabs) to generate voice‑over audio. Image and video generation: Implement T2I, I2V and T2V workflows based on Wan 2.2 or compatible models, preferably directly via Python (without a GUI), alternatively via ComfyUI. Follow the documented parameters (resolution, number of frames, sampling steps, LoRAs, ControlNets, camera movements). Video post‑processing: Combine generated clips into a finished short, add subtitles, music (optional), hooks and CTA elements. Use a library like moviepy/ffmpeg. Platform integration: Connect to the APIs of YouTube, Instagram, TikTok, Facebook and WhatsApp Bus...

...fashion stills suitable for carousel posts and stories • 10 short Reels/TikToks (vertical, 9–15 s each) cut from fresh renders or motion-lifted via Pika Labs, Runway ML, or similar • A clean prompt & metadata file so I can reproduce the look internally • A written transfer of commercial and derivative rights covering the LoRA and all assets Technical expectations Stable Diffusion XL plus ComfyUI (or an equivalent node workflow) is preferred to keep the process transparent. If you lean on Pika Labs or Runway ML for motion, that’s fine—just document the steps so I can rerun them later. All imagery must pass Instagram, TikTok, and Stripe checks—no borderline NSFW angles or brand clashes. Timeline & collaboration I’m ai...

Hi We are working on a creative AI project that combines storytelling and consistent character illustrations — and I’m looking for someone experienced with ComfyUI + Flux + LoRA + InstantID/IPAdapter workflows. we want to build an internal tool/workflow that allows us to create some scenes for story books. spcefically we would need 2 features like has : the "photo to cartoon" feature, where you upload an image of a real person and that person is transformed into a Pixar style character for example (this one should have been done using flux kontext) the "multi character" feature, which we would also need help int reacreting the same outputs as , they managed to do it by @tagging the two different subject and having more control on the scene

Hi We are working on a creative AI project that combines storytelling and consistent character illustrations — and I’m looking for someone experienced with ComfyUI + Flux + LoRA + InstantID/IPAdapter workflows. we want to build an internal tool/workflow that allows us to create some scenes for story books. spcefically we would need 2 features like has : the "photo to cartoon" feature, where you upload an image of a real person and that person is transformed into a Pixar style character for example (this one should have been done using flux kontext) the "multi character" feature, which we would also need help int reacreting the same outputs as , they managed to do it by @tagging the two different subject and having more control on the scene

I already run ComfyUI on RunPod and have a complex, mostly working graph, but I now need to fold in a brand-new module that lets me adjust facial expressions. The job is purely about integrating this new capability into the existing workflow, not rebuilding everything from scratch. What I want inside the graph is clear: • Real-time expression changes • Batch processing of images • Custom expression templates You’ll be working inside the RunPod environment, so part of the task is making sure every node, custom extension and model path resolves cleanly there. If a bespoke node or Python script is required to drive the expression edits, we will wire it in together and confirm it survives container restarts. Deliverables • Updated ComfyUI workflow fi...

I need a web-based user interface developed in HTML, CSS, and JavaScript to connect with the ComfyUI API. The primary functionality required is enabling users to send inputs or commands to the API. Specifically, these commands should be managed through a drag-and-drop interface for intuitive interaction. The interface should be clean, functional, and designed to enhance user experience. Please ensure the UI effectively communicates with the API and provides seamless interaction for the end-users.

Create a ComfyUI workflow (JSON) that generates consistent character poses using ControlNet, supporting both text/image-based character input and style models (anime, pixel art, etc.). Ensure pose accuracy and character consistency across outputs, with options for sprite sheet or separate pose images, and include full dependency installation documentation.

I need assistance downloading and setting up ComfyUI to create AI models. Although I’ve purchased a course, it’s challenging for me to follow, and I’d appreciate guidance to get everything up and running smoothly on my system. This is for personal use, and I need it done as soon as possible. Please highlight your experience with similar setups when applying.

Swap the character out and change voices

...uploads (mobile-friendly) Auto-sequence and animate uploaded images Integrate with open AI text-to-speech for voiceovers (ElevenLabs, Polly, or similar) Add intro/outro templates (logo, contact info, brand colors) Render videos (1 min or less) ready for Reels/YouTube/TikTok Provide basic backend for job management and storage Tech stack (flexible): Python / Node.js / TypeScript FFmpeg or ComfyUI for video rendering OpenAI / Runway / Pika / API for image-to-video generation AWS / Firebase for storage & hosting...

...reproducibility. Deliver watermarked + clean versions of content. Transfer full ownership & rights (no resale/reuse). --- Deliverables: 1 trained LoRA / custom embedding file (.safetensors). 100+ photorealistic influencer-style images (organized by theme). Banner + profile-ready assets. Prompt + settings documentation. --- Requirements: Proven experience with Stable Diffusion (Automatic1111, ComfyUI, or similar). Portfolio with photorealistic AI women / influencer-style outputs. Ability to deliver consistent characters across outfits/scenes. Familiar with post-processing tools (Topaz, Remini, Photoshop) for realism. Willing to sign NDA & Ownership Agreement. --- Budget: Fixed price: $100–$600 (LoRA training + 100–150 images). Ongoing w...

I’m looking for a clean, reusable ComfyUI workflow that lets me go from prompt/ image to image to finished sprite sheet in one smooth run. The core requirements are: • Input flexibility – the node chain must accept (a) a plain text prompt, (b) a reference image, or (c) both together and blend them intelligently. • Pose control – I’ll upload my own line-art or sketch files; the workflow should read those in, keep the structure, and generate the final render around them. • One-shot output – the result I need is a single sprite sheet that lays out every generated pose in the same image. Adjustable cell sizing and sheet resolution are a plus, but the default should “just work” for common game-ready dimensions. • Style...

At Grip (), we are seeking a highly creative and technically adept AI Workflow Specialist, experienced in ComfyUI, to join our team! In this role, you will harness the power of various generative AI technologies to create visually excellent and unique content. Your expertise will be key in creating and configuring AI templates and reusable elements, primarily using ComfyUI and ControlNet, to produce highly controlled and precise visual outputs. This role offers a unique opportunity to combine your passion for generative AI technologies with your talent for creating visually stunning content. You will be working with a team dedicated to pushing the boundaries of what is possible, in creating template-based content generation for the largest CPG brands in the world. If you are

...environment on my new laptop. I want a turnkey installation so I can run Stable Diffusion / ComfyUI for high‑quality imaging and longer video generation, with the flexibility to switch to cloud GPUs (RunPod or Lambda Labs) when I need more VRAM for extended or cinematic projects. Important: Setup access must be by remote session. I will provide a temporary Windows admin account for privacy. System Specs: Laptop: ASUS ROG Zephyrus M16 (GU603Z) CPU: Intel Core i9‑12900H GPU: NVIDIA RTX 3070 Ti Laptop GPU (8 GB dedicated VRAM + shared 8 GB) RAM: [insert confirmed total: 24 GB or 40 GB] DDR5‑4800 Storage: 1 TB PCIe 4.0 SSD OS: Windows 11 Home 64‑bit Deliverables: 1) Install and configure ComfyUI (or Automatic1111 if you recommend it). 2) Add Stable Diffusion XL and ...

I already run ComfyUI on RunPod and have a complex, mostly working graph, but I now need to fold in a brand-new module that lets me adjust facial expressions. The job is purely about integrating this new capability into the existing workflow, not rebuilding everything from scratch. What I want inside the graph is clear: • Real-time expression changes • Batch processing of images • Custom expression templates You’ll be working inside the RunPod environment, so part of the task is making sure every node, custom extension and model path resolves cleanly there. If a bespoke node or Python script is required to drive the expression edits, we will wire it in together and confirm it survives container restarts. Deliverables • Updated ComfyUI wor...

...ratio, resolution, and format for quick upload. Speed matters: I’m aiming for daily drops and several hundred mixed-format pieces per month. Skin texture, anatomy, and lighting must feel natural, with no tell-tale AI artefacts or watermarks. Deliverables • A reproducible, fine-tuned AI model (checkpoint or LoRA) • Prompt library and any control scripts/workflows you build (Stable Diffusion, ComfyUI, Runway, etc.) • Daily packages of ready-to-post images, reels, and short videos, clearly organised by date • A concise hand-off guide so I can extend the pipeline or add new models later Acceptance criteria: the sample batch you provide must reach a “double-take” level of realism, be free of artefacts, and maintain character consistency...

I need a full ComfyUI workflow that lets shoppers virtually try on clothing in real time through an Augmented Reality interface. The system has to handle three garment categories—tops, bottoms and dresses—so the end user can mix and match outfits on-screen and instantly see how each piece drapes and scales to their body. Here’s what I expect: • A clean, well-commented ComfyUI graph that imports a user’s camera feed, isolates the person, and overlays each garment in AR with proper occlusion and lighting. • Modular nodes for the three clothing types, making it simple for me to swap in new textures or 3-D assets without rewriting the flow. • Clear instructions on required dependencies (e.g., OpenCV, ARCore/ARKit bridges, model format...

ComfyUI is already installed on my Windows machine, yet the workflow feels sluggish and the interface is still the vanilla layout. The goal is to tune everything so generating videos for my personal projects feels quick, stable, and visually pleasant. You would diagnose where the current bottlenecks sit (GPU/VRAM usage, queue settings, model-loading order, etc.), apply the right ComfyUI and Python flags, adjust any CUDA / driver parameters if needed, and then tidy up the UI—custom node groups, colour themes, and a cleaner panel arrangement. I’d like the final setup saved in configuration files so I can back it up or replicate it later. Deliverables • An optimised ComfyUI configuration ready to run on Windows without slowdowns or crashes • A ...

ComfyUI is already installed on my Windows machine, yet the workflow feels sluggish and the interface is still the vanilla layout. The goal is to tune everything so generating videos for my personal projects feels quick, stable, and visually pleasant. You would diagnose where the current bottlenecks sit (GPU/VRAM usage, queue settings, model-loading order, etc.), apply the right ComfyUI and Python flags, adjust any CUDA / driver parameters if needed, and then tidy up the UI—custom node groups, colour themes, and a cleaner panel arrangement. I’d like the final setup saved in configuration files so I can back it up or replicate it later. Deliverables • An optimised ComfyUI configuration ready to run on Windows without slowdowns or crashes • A ...

...that can reproduce the subject’s entire body in a photorealistic style inside Stable Diffusion XL. The file will feed a ComfyUI workflow. What I expect to receive • A LoRA (.safetensors) fully compatible with SDXL 1.0 that captures the subject’s likeness and full-length proportions from multiple angles. • A brief note on the training settings (rank, learning rate, steps, resolution, augmentation tricks) so I can fine-tune later if needed. • A handful of sample outputs demonstrating front, side, and three-quarter views without artefacts or drifting likeness. Any LoRA workflow is fine—DreamBooth, LyCORIS, kohya_ss—so long as the final file loads smoothly in ComfyUI. Feel free to use smart augmentation or inpainting to overcome t...