Répertoire des ressources
Trouvez ce dont vous avez besoin pour votre flux de travail.
Guides: techniques, setup, and troubleshootingRecherchez Sweet Tea, Pipes, des flux de travail externes, des modèles, des packs de nœuds et des guides dans un catalogue unique lié aux preuves.
Commencez plutôt avec un flux de travail ↗Résultats de la recherche
Toutes les ressources
20 affiché of 42 619
FLUX.1 Krea image generation
Create images from text using the fixed FLUX.1 Krea dev model, CLIP-L and T5 text encoders, and VAE. No uploaded image is required. For a first run, describe your subject, composition, lighting, and style in the positive prompt. Keep the default 1024 × 1024 size, one image, 20 steps, Euler sampler, and Simple scheduler. The supplied portrait prompt is an example; replace it for another subject. Width and height are measured in pixels and must be multiples of 16. Image count generates a batch with the same prompt and dimensions. Larger dimensions and image counts require more memory and computation. Open Sampling for advanced controls. Seed defaults to -1, which randomizes each run. Use a fixed seed when comparing prompt or sampling changes with other settings unchanged. Steps controls the number of denoising steps; sampler selects the sampling algorithm, and scheduler distributes noise levels across those steps. CFG defaults to 1; higher values amplify the difference between positive conditioning and zeroed negative conditioning. This graph has no editable negative prompt or separate FLUX guidance control. Denoise controls the fraction of the sampling schedule used, from 0 to 1. Keep its default of 1 for full generation. The workflow always starts from an empty latent, so reducing denoise does not enable image editing or preserve an existing image. Model and encoder selection are fixed in this tool.
Z-Image upscale and refine
Upload an image to produce one resized, RealESRGAN-upscaled and Z-Image Turbo diffusion-refined image. The selected result includes the entire processing chain; there is no separate upscale-only output or refinement bypass. For a first run, upload your source image and describe its subject, composition and desired details in the positive prompt. Keep the source defaults: 1 megapixel working size, 0.5 final scaling, 0.33 refinement strength, five sampling steps and prompt guidance (CFG) 1. The default prompt, “masterpiece, 8k,” is a minimal starting point; its text does not determine resolution. Working size sets the pixel budget before the fixed 4× RealESRGAN upscaler. Final scaling then multiplies the upscaled width and height before diffusion refinement. Approximate output megapixels equal working megapixels × (4 × final scaling)². The defaults therefore produce approximately 4 megapixels. Aspect ratio follows the upload, subject to image and latent rounding. This is not a fixed enlargement relative to the original upload; the working-size stage can also shrink a large source. Refinement strength controls how much diffusion can reinterpret the resized image. Lower values maintain more source structure; higher values allow stronger changes. Use a prompt that describes the image and the changes you want. Refinement can alter details rather than merely recover them. The negative prompt has no effect at CFG 1. It becomes useful for describing unwanted attributes when experimenting with guidance above 1. Steps, sampler and scheduler change refinement behavior. Advanced model sampling shift is a separate noise-schedule control, retained at the source default of 3 for the first run. The model setup is fixed to Z-Image Turbo with the Qwen text encoder, ae VAE and RealESRGAN_x4plus upscaler. Seed -1 randomizes each run. Use a fixed seed to compare settings with the same noise initialization. The Pipe uses the loader's RGB image output and does not preserve uploaded transparency.
Create a HeyGen talking-head video
Generate one talking-head video with a HeyGen avatar speaking your written script. For a first run, replace the demo script, keep Annie Lounge Standing Side and (avatar's default voice), and leave resolution at 720p, aspect ratio at auto, and speech speed at 1. Avatar and voice are text fields for supported HeyGen selections; this Pipe does not supply selection lists. Custom avatar and voice ID fields accept existing HeyGen identifiers. Leave them empty when using the regular selections. These fields do not create or train an avatar or voice. Choose 720p or 1080p resolution and an automatic, landscape, portrait, square, 4:5, or 5:4 aspect ratio. Background color accepts a HeyGen-supported color value; leave it empty for the provider's default behavior. Speech speed is a multiplier from 0.5 to 1.5: 1 is normal speed, lower values slow speech, and higher values accelerate it. The advanced Video output section offers automatic, MP4, MKV, or WebM file type and automatic, H.264, or AV1 codec selection. Leave both on auto for a first run. Seed -1 randomizes each run; a fixed seed does not guarantee identical provider-generated video. This retained workflow uses the automatic avatar engine and script speech only. It has no audio upload, reference-image input, separate language selector, or selectable alternative engine or speech branch.
Kandinsky 5 text to video
Generate a short silent video from a scene and motion description using Kandinsky 5 Lite's fixed 5-second text-to-video model. For a first run, replace the example positive prompt with your subject, action, setting, camera movement, lighting and visual style. Optionally enter unwanted content or visual traits in the negative prompt, or leave it blank. No image or audio upload is required or connected. Keep 768 × 512 pixels, 121 frames, 24 FPS, batch size 1 and the supplied sampling settings: 50 steps, CFG 5, Euler ancestral sampler, beta scheduler, sampling shift 5 and denoise 1. These defaults produce approximately five seconds of video, returned as an H.264 MP4. Seed -1 randomizes each run; use a fixed seed when comparing changes. Width and height use multiples of 16. Frame count uses the 4n + 1 pattern, such as 121. FPS sets playback timing: approximate duration in seconds is frame count divided by FPS. Changing FPS plays the same generated frames faster or slower; it does not generate additional frames. Larger dimensions and frame counts increase computation and memory requirements. Changing frame count does not select a different model duration variant. Advanced sampling controls change denoising behavior and prompt guidance. Negative-prompt influence depends on CFG. Keep denoise at 1 initially: this workflow starts from an empty latent, so lowering denoise does not preserve an uploaded image. Batch size controls how many latent sequences are generated together, but the output path assembles one video and provides no separate clip outputs for batch members. Video encoding offers auto, MP4, MKV and WebM containers, with auto, H.264 or AV1 codecs. Keep MP4 with H.264 initially; other combinations depend on output transport support. Quality (CRF) defaults to -1, which uses the transport's default encoder setting. Encoding controls affect the returned file rather than the generated scene. This Pipe has no connected image animation or audio mode.
Wan Animate 2: Motion transfer
Animate a reference character using motion from a driving video, with text control over appearance and background. For a first run, upload a character image and a driving video with similar subject framing. Replace the appearance example with a description of your character and desired background, and replace the motion example with the action shown in your video. Leave the sampling defaults and context windows off initially, then generate the animation. The default requests one 81-frame segment at 482 × 854 pixels, using the driving video's frame rate and audio when present. Width and height resize the driving frames; the reference image is resized to the resulting dimensions. Both use center cropping for aspect-ratio mismatches. Set one dimension to 0 to derive it from the other using the driving video's aspect ratio; do not set both to 0. The reference upload's alpha mask is not connected, so uploaded transparency is not preserved. Frame count sets the requested segment length. Motion strength and reference strength independently control their respective conditioning. Motion guidance start and end specify the sampling interval for pose guidance, with 0 to 1 covering the full process. Starting frame offsets motion conditioning, but the driving video's first frame still supplies its vision conditioning and the attached audio is not offset. Audio synchronization is therefore not guaranteed when changing the starting frame or removing the first generated frame. Optional context windows switch to temporal-window sampling. Window length, overlap, schedule, blending and FreeNoise apply only when enabled. Stride applies only to uniform schedules, and closed loop only to looped schedules; this toggle does not guarantee a seamless video loop. The schedule and blending fields retain source names without an enumerated list of alternatives, so use the defaults unless you know another value supported by the installed node. Advanced sampling retains the source's Lightx2v distillation LoRA at strength 1, six steps, lcm sampler, simple scheduler, CFG 1 and sampling shift 5. Seed -1 randomizes noise; a fixed nonnegative seed reuses its initialization. Denoise, added noise, latent batch size and LoRA strength remain available for advanced changes. Batch size defaults to 1; this Pipe returns one video rather than separate outputs for batch members. Output container and codec default to automatic selection, and CRF -1 uses the transport default. The delivered result is the generated animation. The source's side-by-side comparison is auxiliary, and its disconnected continuation stage does not extend this output.
Z-Image Turbo text to image
Create images from a text description using the fixed Z-Image Turbo INT8 model, Qwen text encoder and paired VAE. No uploaded image is required. For a first run, replace the example prompt with your subject, composition, lighting and style. Keep the supplied 1024 × 1024 dimensions, one image, eight steps and CFG 1, then run the tool. Width and height are separate pixel dimensions in multiples of 16 and determine the actual aspect ratio; composition wording in the prompt does not change them. Increase image count to return a batch, with greater memory and processing requirements. Seed -1 selects new random starting noise each run. Set a fixed seed when comparing prompt or sampling changes. In advanced Sampling, the sampler selects the denoising algorithm and the scheduler sets the noise levels across steps. Start with res_multistep and simple. CFG 1 uses positive conditioning without classifier-free amplification; higher values amplify guidance against zeroed negative conditioning. There is no editable negative prompt. Denoising fraction controls how much of the sampling schedule is used. Keep its supplied value of 1 for the initial run. Noise schedule exposes the model sampling shift, initially 3, which changes the schedule before denoising. These controls interact with the sampler, scheduler and steps; a fixed seed helps compare their effects. This tool generates from an empty latent and cannot transform an uploaded image. The model, text encoder and VAE are fixed. Every decoded image in the batch is returned to Sweet Tea; the selected output stage does not permanently save images through ComfyUI SaveImage.
Text to audio with Stable Audio Open
Generate music or sound from a text description with Stable Audio Open and export it as MP3. For a first run, keep the example audio prompt or replace it with a description of your desired sound, instruments, mood, or musical style. Keep the source defaults: 47.6 seconds, one audio sample, and V0 MP3 encoding. Run the Pipe to generate and export the audio. No media upload is required. The workflow requires stable-audio-open-1.0.safetensors in the checkpoints model directory and t5-base.safetensors in the text_encoders model directory. Both are identified by the source; the supplied dependency records mark them as missing, so they must be available before execution. Duration sets the requested length in seconds. Audio count generates a batch of samples with the same prompts and settings. MP3 quality selects V0, 128k, or 320k export encoding. In Advanced, leave the negative prompt empty or describe unwanted qualities. Seed -1 randomizes generation; a fixed seed reuses the starting noise for comparisons with otherwise unchanged settings. Advanced sampling controls expose the source's denoising steps, prompt guidance, sampler, noise schedule, and denoising strength. Start with the supplied settings. The workflow starts from an empty audio latent and does not accept existing audio for editing. The duration widget's broad range and the sampler menu do not establish model support for every length or alternative algorithm. Changing MP3 encoding affects export, rather than the generated content.
Vidu Q1 text to video
Generate one five-second 1080p video from a text prompt with Vidu Q1. For a first run, use the supplied snowboarding prompt or replace it with a description of your subject, setting, action, and camera framing. Keep aspect ratio at 16:9 and movement amplitude at auto, then run. No uploaded image is required. Choose 16:9 for landscape, 9:16 for portrait, or 1:1 for square framing. Movement amplitude requests small, medium, or large object movement; auto lets the service choose. Duration is fixed at five seconds and resolution at 1080p in this configuration. Camera framing is described through the prompt. Use seed -1 to randomize each run, or enter a nonnegative seed to request a particular variation. Identical results across provider runs are not guaranteed. The advanced Video output stage offers file type and codec choices. Leave both at auto for automatic encoding selection. The generated video is returned to Sweet Tea through its preview transport. Generation sends your prompt to the Vidu partner API and requires an authenticated runtime with sufficient credits and service access.
Wan 2.1 transparent animated WebP
Generate a transparent animated WebP from a text description using Wan 2.1 and its RGBA adaptation. Color and alpha are decoded from the same generated video latents and combined before encoding. No uploaded image is required. For a first run, replace the example prompt with your subject, movement, framing, and visual style, and request a transparent background. Keep 832 × 480 pixels, 33 frames, batch size 1, and 16 playback FPS for approximately 2.06 seconds. Width and height use multiples of 16; frame counts follow 1, 5, 9, and so on. Larger dimensions and more frames increase generation work. FPS changes playback speed and duration without generating more frames. Advanced Sampling controls include Seed, Steps, CFG, sampler, scheduler, and denoising amount. Seed -1 randomizes each run; use a fixed seed to compare settings. The default configuration uses 20 steps, CFG 6, uni_pc, simple scheduling, and full denoising. The negative prompt lists unwanted content and artifacts; its supplied default is in Chinese. Lightx2v is optional and bypassed by default. To use the documented accelerated configuration, enable that stage and manually set Steps to 4 and CFG to 1. Its strength has no effect while bypassed. Keep RGBA adaptation strength 1 and sampling shift 5 for a first run. RGBA adaptation strength changes the model, not output opacity. WebP output exposes playback FPS and encoding controls. Lossless encoding is enabled by default. Disable it for lossy encoding, where Quality controls detail retention; in lossless mode Quality controls compression effort. Keep Compression method at default initially. Keep batch size 1 for one animation: this selected output encodes the resulting image batch into animated WebP and does not provide separate video outputs. The result is animated WebP rather than MP4. Use a viewer or compositing tool that supports animated WebP and alpha. Generated subject boundaries and fine details may need review before compositing.
Ideogram 4 image generator
Generate images from scene prompts using the fixed Ideogram 4 conditional and unconditional model pair. Enter your completed scene description in Positive prompt; structured JSON is recommended for compositions containing text and positioned objects. No uploaded media is required. For a first run, keep the supplied monochrome knight-and-horse magazine-cover prompt, square aspect ratio, 1 megapixel output, Default sampling preset and Image count 1, then generate. The example mentions image_0f5a20.png, but there is no reference-image input and that file is not supplied to the model. Replace that mention with a written typography description when adapting the example. Technical configuration text inside the prompt is requested lettering in the image; it does not change the actual sampler, steps or seed. Output size controls both the latent canvas and scheduler dimensions. Dimensions round upward to multiples of 16, with a minimum of 256 pixels per side. Changing the canvas aspect ratio does not rewrite the prompt's JSON aspect_ratio or bounding boxes; adjust those manually to suit your composition. Optional boxes use [y1,x1,y2,x2] coordinates normalized to 0–1000. Sampling preset selects coordinated scheduler settings: Default uses 20 steps, mu 0 and std 1.75; Quality uses 48 steps, mu 0 and std 1.5; Turbo uses 12 steps, mu 0.5 and std 1.75. These names identify sampling configurations rather than guaranteed output quality. Advanced controls expose guidance, a guidance override interval, seed, sampler and image count. Main CFG defaults to 7; Override CFG defaults to 3 between progress 0.7 and 1.0. Keep Override start below Override end. Set the override equal to main CFG to remove their guidance difference. Seed -1 selects fresh noise each run; a fixed seed allows comparisons with the same noise initialization. Euler is the source sampler default. Larger image counts and resolutions require more memory and computation. The workflow uses ideogram4_int8_convrot.safetensors and ideogram4_unconditional_int8_convrot.safetensors in diffusion_models, qwen3vl_8b_fp8_scaled.safetensors in text_encoders, and flux2-vae.safetensors in vae. These fixed models must be available in the execution environment. The supplied text-encoder installation source is https://huggingface.co/Comfy-Org/Qwen3-VL/resolve/main/text_encoders/qwen3vl_8b_fp8_scaled.safetensors. There is no independent negative prompt: negative conditioning is zeroed from positive conditioning and passed to the unconditional model. The source's separate external-LLM instruction template is outside this image output. This Pipe does not call an LLM or automatically turn an idea into JSON; paste your finished prompt directly into Positive prompt.
SeedVR2 image upscale
Create an enlarged, restored image with SeedVR2 and optional source color matching. Upload the image to restore, leave Scale factor at 4 and Color matching at none, then run. The default multiplies both dimensions by 4, producing approximately 16 times the source pixel count before restoration. No text prompt is required: the uploaded image supplies conditioning. The model is fixed to SeedVR2 3B int8 with its matching VAE. Scale factor controls the initial resize: values above 1 enlarge, 1 retains the source dimensions and values below 1 reduce them. SeedVR2 restoration still runs in each case. Resampling method controls interpolation before restoration; lanczos is the source default for enlargement, area supports reduction and nearest-exact retains hard pixel boundaries during resizing. Restoration may subsequently change those details. Color matching uses the resized source as its reference. Choose none to skip color transfer, lab for CIELAB color transfer, wavelet to transfer broad color while retaining restored high-frequency detail, or adain to match per-channel mean and standard deviation. Advanced restoration settings include Seed, Denoise, Steps, Conditioning guidance, Sampler and Scheduler. Seed -1 randomizes each run; use a fixed seed to keep the sampling noise consistent when comparing settings. For a first run, retain denoise 1, one step, guidance 1, Euler sampling and the simple schedule. Other sampling configurations are experimental for this source setup. Advanced input and output tiling independently control spatial encoding and decoding. Tile size and overlap are in pixels; keep each overlap below its corresponding tile size. Larger output dimensions increase the amount of image data processed. Restoration can reinterpret fine details rather than recover exact missing information. Uploaded image and alpha-mask outputs are joined before resizing and preprocessing, but final output transparency is not established; do not rely on this tool to preserve transparency.
Gemini reference images to video
Generate one 720p, 24 FPS video with audio from two reference images and a scene or story prompt using Gemini Omni Flash. For a first run, upload both required images: a primary character, scene or style reference as Reference image 1, and a storyboard or complementary reference as Reference image 2. Replace the example prompt with your own action, visual style, camera behavior and desired sound. Explain how each image should guide the result. For a storyboard, describe panel reading order and the action to animate. Use 9:16 for portrait or 16:9 for landscape, and a requested duration from 3s through 10s. The defaults request a 3-second portrait clip. These values are appended to the prompt as provider instructions; avoid conflicting duration or aspect-ratio instructions in your main prompt. Both images are general references, without guaranteed start-frame or end-frame roles. Only their image outputs are supplied; uploaded alpha masks are not passed to generation. Generation tuning contains Temperature, Top P and Seed. Temperature controls sampling randomness; Top P controls the cumulative probability threshold for nucleus sampling. Keep their defaults for a first run. Seed -1 randomizes the re-run seed each run. Seed affects re-running and caching; a fixed seed does not reproduce the same video. Video output offers optional file type, codec and encoding-quality settings. Leave file type and codec on auto and CRF at -1 for the first run. These settings affect output encoding, rather than the provider's creative generation. Generation sends your prompt and reference images to a hosted Gemini service, requires configured provider access and uses provider billing.
Z-Image Turbo with PiD 4× upscale
Generate a 4096 × 4096 square image from text with Z-Image Turbo and PiD, and receive the 1024 × 1024 base preview alongside it. The same prompt guides both stages; PiD uses the generated base latent as conditioning for its final pixel-space image. For a first run, replace the example prompt with your subject, composition, lighting and style, then keep the supplied defaults: eight base steps and four PiD steps, guidance of 1 for both stages, res_multistep base sampling and LCM PiD sampling with simple schedules. No media upload is required or supported. The base and PiD seeds control separate stages. Use -1 to randomize each run, or fix both seeds when comparing settings. Changes to base sampling affect both the base preview and the latent that conditions PiD. PiD sampling controls affect the final image. Advanced settings expose each stage's sampler, guidance and schedule, plus base sampling shift, denoising fractions and PiD noise initialization. Keep both denoising fractions at 1 and Add PiD noise enabled for the supplied setup. Disabling PiD noise initialization removes the PiD seed's effect on initial noise. Latent degradation is separate from denoising fraction: zero denotes a clean base latent, while higher values tell PiD to handle corrupted latent outputs. The model pair, square dimensions and single-image batches are fixed for this 1024-to-4096 setup. Both stages run to produce the final result; there is no base-only or uploaded-image upscale mode. There is no editable negative prompt: both stages use zeroed negative conditioning. Alternative sampling settings change the generation behavior without a guaranteed quality improvement.
Magazine cover
Create a magazine-cover image from a reference photo or a written description. For a first run, upload a photo, replace the masthead and cover lines in Cover prompt, and keep the portrait 3:4 aspect ratio and 2K resolution. To generate from text alone, disable the Source image stage and describe the subject, background and typography in Cover prompt. Choose an explicit aspect ratio for text-only generation. The default recipe requests flat, minimalist editorial typography: a large pale-yellow masthead, small white cover lines, asymmetric placement and clean negative space. Replace the example wording with your own text. For a street-poster border collage, request cropped lettering, rotated microtext and symbols around the edges while keeping the center and subjects clear. For a foreground-subject variation, explicitly ask for the subjects to appear in front of the typography. Apply these recipes by editing Cover prompt; they are generative instructions rather than separate stages or editable layers. Model selects between gemini-3-pro-image-preview and Nano Banana 2 (Gemini 3.1 Flash Image). Aspect ratio sets the cover's width-to-height proportions; auto follows a supplied reference image. Resolution offers 1K, 2K and 4K, with the provider's native upscaler used for 2K and 4K. Seed -1 randomizes each run; a fixed seed requests best-effort consistency without guaranteeing identical results. The result is a generated image rather than editable text or layout layers. Review spelling, text placement, subject preservation and changes to the reference photo before use. Only the image is returned; requesting IMAGE+TEXT does not add a text output. This tool returns the cover directly without running the separate packaging-presentation stage.
Swap product into a reference ad
Generate an advertisement featuring your product, using an existing ad as the composition and style reference. For a first run, upload a reference advertisement and a clear image of your replacement product. Replace the example Product and copy brief entirely with your product description, campaign intent, desired colors, headline, and supporting copy. Keep aspect ratio on auto and the defaults of 4K resolution and gemini-3-pro-image-preview, then run and inspect the returned image. The reference ad is supplied first for composition, lighting, graphic treatment, and text layout. The product image is resized and padded to the reference dimensions and supplied second. The editing template asks the model to replace the featured product and harmonize lighting and texture. It requests replacement of existing advertising copy while retaining its styling; if the reference has no advertising copy, it requests no added text. Text physically on the product is treated separately from advertising copy. Model selects gemini-3-pro-image-preview or Nano Banana 2 (Gemini 3.1 Flash Image). Resolution offers 1K, 2K, and 4K; the source describes 2K and 4K as using the provider's native upscaler. Auto follows the reference proportions. Explicit aspect ratios request a different output shape, but the editing template also requests the reference ratio, so these instructions can conflict. Under Product reference preparation, choose white or black padding and the product resize method. These affect the product reference supplied to the model; padding color does not directly set the final ad background. Keep the default area resize method for a first run. Uploaded alpha masks are not connected, so these inputs do not preserve transparency or supply a replacement mask. Seed -1 requests a new random seed each run. A fixed seed requests best-effort consistency and does not guarantee identical results. Provider response format offers IMAGE or IMAGE+TEXT; only the generated image is delivered by this Pipe, even when accompanying provider text is requested. This is a generated reinterpretation rather than a precise masked composite. Layout, lettering, branding, requested wording, and product fidelity can vary. Review the result before using it.
Qwen Image Edit
Create an edited version of an uploaded image with Qwen Image Edit. Upload your source image, then describe what should change and what should stay the same. Replace the supplied underwater-scene example with instructions for your own image. For a first run, leave Lightning mode off and use the defaults: 20 standard steps, CFG 2.5, Euler sampler, simple scheduler and Denoise 1. Lightning mode selects the fixed four-step Lightning LoRA together with its separate steps and CFG controls, initially 4 and 1. Only the steps and CFG for the selected mode affect the result. Lightning LoRA strength affects only Lightning mode; guidance normalization and sampling shift apply in both modes. Their supplied defaults are 1 for LoRA strength, 1 for normalization strength, normalization before CFG off, and sampling shift 3. Denoise ranges from 0 to 1: lower values retain more source structure, while 1 allows stronger regeneration. Negative prompt describes unwanted content and may remain empty. Seed -1 chooses a new random seed each run; use a fixed seed to compare settings with the same initial noise. Advanced controls expose the sampler, scheduler and separate sampling settings for each mode. Step counts must be positive integers; the downstream sampler supports 1–10000 steps and CFG values from 0 to 100. The source image supplies both visual conditioning and the starting latent. There is no connected output-size control or automatic 1.5-megapixel resize; resize uploads beforehand if you need a different starting size. The uploaded alpha mask is unused, so source transparency is not preserved. There is no mask control to restrict the edit to a selected area. Editing is generative and may change details beyond the requested changes.
Dieline to box mockup
Create a generated packaging mockup from a flat dieline image and a rendering prompt. For a first run, upload a dieline with its artwork and panel layout visible, leave the Dieline image stage enabled, and run the supplied prompt. It requests an upright box in a three-quarter view on a neutral studio background. Edit Rendering prompt to change the packaging form, camera angle, lighting, or setting. The defaults use Gemini 3 Pro Image, automatic aspect ratio, and 4K target resolution. Auto follows the reference image's proportions; choose an explicit ratio when the flat dieline's shape is unsuitable for the finished composition. The 2K and 4K options use Gemini's native upscaler. Model selects between Gemini 3 Pro Image and Nano Banana 2; their interpretations can differ. Seed -1 requests a new random seed each run. A fixed seed requests best-effort consistency, without guaranteeing identical results. System instructions set general model behavior; the supplied instructions request image generation. For text-only packaging concepts, disable the Dieline image stage and rewrite the prompt to describe the desired packaging and artwork without referring to an uploaded dieline. No upload is needed in this mode. Choose an explicit aspect ratio for a predictable composition shape. Generation uses the Gemini provider and sends the prompt and enabled reference image to that service through a paid provider call. The selected result is the generated image, returned through the Pipe's preview output; a text response is not included. This produces an image, not a dimensional 3D model or a verified folding simulation. Inspect panel placement, artwork, and typography: exact reproduction is not guaranteed, even though the supplied prompt requests it.
Apoc, Lost, Decay and Ruin - What a perfect day to end the world (Apocalypse Embed Series for 2.1)
As a fan of post apocalyptic series like The Mountain Man or Arisen , and of course great video games like The Last of Us and Project Zomboid, I really wanted an easy way to recreate that "world's gone to hell and we...
SimpleNegative | Negative TI
I created 'SimpleNegative' because there are too many repetitive responses influenced by 'EasyNegative,' and it's becoming boring. Please place it in NegativePrompt in the same way.
Photo to geometry maps
Upload one perspective photo to produce two geometry-map images from a shared MoGe-2 estimate. For a first run, upload your image, keep the defaults, and run the Pipe. The first image shows OpenGL normals and the second shows DirectX normals, with inference detail set to 9, automatic camera field of view, and invalid-region filtering enabled. Choose each map independently: OpenGL normals, DirectX normals, depth, colored depth, or the predicted validity mask. OpenGL normals use green = +Y; DirectX normals use green = −Y. Both outputs use the same geometry estimate and inference settings. In Geometry estimation, inference detail ranges from 0 (fastest) to 9 (most detail), as described by the source. Enter the camera's horizontal field of view in degrees when known, or leave 0 to recover it automatically from predicted points. This controls the focal length used to unproject depth into 3D. Invalid-region filtering marks predicted sky and invalid pixels as invalid in points and depth; disable it to retain raw predictions. The predicted mask remains selectable either way. Uploaded transparency is not used as a mask. Automatic resizing checks image width: images wider than 2048 pixels are resized to a 2048-pixel longer edge. Images whose width is at most 2048 remain unchanged, including tall images. Disable automatic resizing to use the original image size. Inference batch size specifies images per inference call, rather than output variations; this Pipe supplies one photo. Force projection is enabled by default, but the supplied source does not explain its effect on the maps. These images are estimates from visible photo content, not guaranteed accurate measurements. This Pipe returns image maps and does not export a mesh or GLB. It requires the fixed moge_2_vitl_normal_fp16.safetensors checkpoint under models/geometry_estimation/.
