Encuentra lo que necesita tu flujo.

ModeloCivitai

bad_pictures

A very versatile negative embedding that helps generate high quality images. Higher chances of producing normal looking hands in realistic images.

TextualInversionSD 1.5

FLUX.2 [klein] 4B1024x10248GB VRAM

Flux 2 Klein four-step text to image

Create images from a text description using the fixed distilled Flux 2 Klein 4B model. For a first run, replace the example Positive prompt with your subject, setting, composition, lighting and desired visual style. No uploaded image is required. Keep the source defaults: 1024 × 1024 pixels, one image, four steps, CFG 1 and Euler sampling. Run the Pipe to receive the generated image in Sweet Tea. Width and height set both the image canvas and the resolution used by the sampling schedule. Enter positive multiples of 16, at least 16 pixels. Together they determine the aspect ratio. Images per run generates a batch with the same prompt and dimensions; larger dimensions and batches require more memory. Seed -1 selects fresh random starting noise each run. Use a fixed nonnegative seed to compare changes with the same starting noise. Advanced controls expose steps, CFG and sampler. Steps set the sampling iteration count; additional steps add computation and do not guarantee better results. Keep four steps and Euler for the source setup. Changing the sampler can change results and interact with the step count. CFG adjusts positive conditioning relative to zeroed conditioning. Keep 1 for the source setup; higher values change that guidance. This branch has no editable negative prompt. This Pipe generates only the distilled-model branch's images. The separate Klein Base model with editable negative prompting belongs to the sibling tool and is not a selectable mode here. The runtime requires flux-2-klein-4b.safetensors in diffusion_models, qwen_3_4b.safetensors in text_encoders, and flux2-vae.safetensors in vae. These model dependencies are fixed. The output transport returns images to Sweet Tea without permanently saving them through ComfyUI.

Kling motion transfer

Animate a character image using movement and expressions from a reference video, returning one generated video. For a first run, upload a character image and a 3–10 second motion reference video. Keep Orientation source set to video, replace the example Scene prompt with your character, setting and desired details, then run the Pipe. With video orientation, movement, expressions, camera motion and character orientation follow the reference video; supported reference duration is 3–30 seconds. With image orientation, movement and expressions still follow the video, character orientation follows the image, and camera details come from your prompt; supported reference duration is 3–10 seconds. Keep reference audio determines whether the motion video's original sound is retained and is enabled by default. Generation mode offers pro and std, while Model version offers kling-v3 and kling-v2-6. Defaults are pro and kling-v3; the supplied source does not establish quality or performance differences between these choices. Advanced Video output settings let you choose a file type and codec; both default to automatic selection. The CRF encoding control defaults to -1, and its codec-dependent behavior is not documented in the supplied evidence. Generation uses the Kling provider and transfers the reference media for a provider-priced request. Uploaded image transparency is not preserved because only the loader's IMAGE output is connected.

Styled logo rotation

Generate a styled logo animation from a logo image and a material or matcap reference. The material reference is resized and padded to the logo's dimensions, then both images guide generation of a styled first frame. Gemini 2.5 Flash Image generates the last frame from that first frame, and Seedance animates between the two generated endpoints. For a first run, upload both required images and keep the default prompts and five-second duration. These request a studio-lit logo, a matching back view, and a 180-degree rotation. Prepare the logo's background before uploading: uploaded transparency is not carried through the connected RGB image inputs. In the styling prompt, image 1 is the logo and image 2 is the material reference. In the last-frame prompt, image 1 is the generated styled first frame. To change the movement, edit the last-frame and motion prompts together so the destination view agrees with the requested motion. Fixed camera adds a stationary-camera instruction; the default motion prompt also explicitly requests a static camera. If you want camera movement, disable Fixed camera and revise that prompt. First-frame model selects gemini-3-pro-image-preview or Nano Banana 2 (Gemini 3.1 Flash Image). First-frame shape and resolution govern the initial image; the last frame defaults to matching its shape. Video shape and resolution govern the final clip independently. Duration is 3–12 seconds. Generate audio applies only to the Seedance 1.5 Pro model. AI watermark requests a watermark from the video service. Reference preparation offers white or black padding and a resampling method. Video export offers container and codec choices; keep their Auto defaults for a first run. Both endpoint images are generated on every run; uploaded endpoint frames are not supported. Each generation stage has its own seed: -1 randomizes that stage. Fixing only the video seed does not fix the generated endpoints. Fixed Gemini seeds request best-effort repeatability rather than deterministic results. Geometry, lettering, material consistency, and camera stability are not guaranteed. The selected result is the video; model text responses are not connected to an output.

Color Palettes

Easy one-click color palette for your images. No need to argue with the AI if an orange is a color or a fruit, everything is setup for you automatically.

TextualInversionSD 1.5

Qwen-Image-Edit1024x10248GB VRAM

FlashVSR video upscale

Upscale an uploaded video to 720p, 1080p, 2K, or 4K with WaveSpeed FlashVSR and return one processed video. For a first run, upload your source video, keep Target resolution at 1080p, and leave File type and Codec on auto. Run the Pipe to receive the upscaled video. The uploaded video is sent to WaveSpeed for processing through a paid provider call. Target resolution controls the FlashVSR request. Under advanced Video output settings, choose auto, MP4, MKV, or WebM for the container and auto, H.264, or AV1 for the codec. These controls affect the returned video's encoding rather than the requested upscale resolution; use automatic encoding unless you need a specific format. This Pipe processes an existing video. It exposes no prompts, restoration strength, frame-rate changes, or duration controls. The source CRF default is retained as a hidden encoding setting because its behavior is not documented in the supplied evidence.

赛博世界/Cyberworld/赛博朋克场景 Lora

推荐的质量词:masterpiece, best quality, score7, score8, score9, highres, extremely detailed 8k wallpaper, illustration, depth of field, great quality, amazing quality, very aesthetic, absurdres, cyberpunk, cyberpunk city,

Qwen Image Edit 2511

Edit an uploaded image with text instructions using Qwen Image Edit 2511. Upload a source image, replace the example instructions with the change you want, and run with the supplied defaults. Leave the negative prompt empty for a first run. The result is the decoded edited image. The uploaded image is automatically scaled and used both as a reference for prompt conditioning and as the initial image latent. There are no independent output-size controls. Denoise ranges from 0 to 1: lower values retain more initial structure, while 1 uses the full denoising schedule. Uploaded transparency is not preserved, and masked editing is not supported. Lightning mode switches together to the fixed four-step Lightning LoRA, Lightning steps and Lightning CFG. Standard mode defaults to 40 steps and CFG 4; Lightning mode defaults to 4 steps and CFG 1. Each mode's steps and CFG controls apply only while that mode is selected. Leave Lightning LoRA strength at 1 for the supplied configuration. Negative prompting has limited influence at CFG 1. Seed -1 selects new random starting noise on each run. Use a fixed seed to compare changes with the same starting noise. Sampler and scheduler apply in both modes; the supplied settings are Euler and simple. Advanced controls also expose guidance normalization strength, normalization placement and sampling shift, all shared by both modes. Normalizing before CFG can amplify noise; the default uses post-CFG attenuation only. Positive and negative reference-latent methods are separate controls for their respective conditioning paths. Keep their supplied index_timestep_zero values for a first run. These settings do not add more reference images or change output dimensions.

Z-Image Turbo1024x10248GB VRAM

Two-pass image generator with tiled enlargement

Generate finished images from text with two Moody Z-Image diffusion passes and tiled enlargement, with optional face refinement and skin-contrast finishing. No uploaded image is required. The Pipe returns the finished image batch through Sweet Tea's image output; it does not promise permanent saving. For a first run, replace Positive prompt, review Negative prompt, and retain the 640 × 960 initial size, 1.7 Intermediate scale and 2× Enlargement factor. Image count determines the number of generated images. Final dimensions are approximately the initial dimensions multiplied by both scales, subject to latent-size rounding. Enlargement factor supports 0.05–4 in increments of 0.05; values below 1 reduce size. Larger dimensions and image batches increase processing and memory requirements. Positive prompt guides both generation passes and, by default, tiled enlargement. Negative prompt affects only the initial pass; subsequent stages use zeroed negative conditioning. Enable Use separate enlargement prompt to replace the main positive prompt during tile redraw only. Review the saved Enlargement prompt before using it because it contains portrait-specific styling. Enlargement denoise controls how much tile redraw can change the generated image. Keep the supplied sampling schedules and boundaries for a first run. Initially, a 17-step schedule runs from step 0 to step 12 and retains leftover noise. After latent enlargement, refinement uses an 11-step schedule starting at step 5 and ending beyond the schedule so it completes. Steps describe full schedule lengths, rather than simply executed iteration counts. Changing a schedule, its boundaries, scheduler or model timing changes the handoff or refinement behavior. Keep each starting boundary below its schedule length and stopping boundary. Each sampling stage has an independent seed; -1 requests fresh randomization. Fix all active stage seeds when comparing settings. Tile dimensions are calculated automatically from the decoded dimensions and Enlargement factor, using a fixed 96-pixel constant in the source formula. Changing Tile padding does not update that constant. Tile batch size controls how many tiles are processed together, rather than how many finished images are returned; it can change results and memory use. Seam fix mode defaults to None. Enabling a supported seam mode activates its separate redraw-strength, band-width, blur and padding controls. Sampler, scheduler and seam-mode choices without listed options depend on the installed node implementation. Automatic face refinement starts disabled. Enable its stage to detect and redraw faces, then replace the separate Face prompt for your subject; it does not inherit the main prompt, and its saved text is portrait-specific. Face denoise controls redraw strength. Detection threshold selects candidate faces, crop and mask controls determine the processed region, and guide and maximum sizes govern region-processing dimensions. Skin contrast also starts disabled. Enable its stage before adjusting Skin contrast strength. With normal blending, 0 keeps the incoming image and 1 uses the full model-processed result. When the stage is disabled, normal blending leaves the image unchanged; other blend modes can still alter tones by combining identical image branches. The diffusion checkpoints are fixed to the source's moody-wild-v4-undistilled-40steps-cfg4_00001__clean.safetensors and moody-real-v6.4_00001_.safetensors. Generic Z-Image dependency entries do not establish equivalent replacements. The retained native editor bridge and manual detailer remain hidden and bypassed; painted-mask refinement is unavailable in this interface.

Fashion editorial from three references

Create a fashion editorial image from separate character, background, and product references using Gemini. For a first run, upload all three required images and keep the example prompt. The references reach the model in this order: image 1 is the character and clothing, image 2 is the background, and image 3 is the product. The example requests a hard-flash, grainy editorial with the character showcasing the product. Edit the prompt's pose, expression, clothing, product placement, lighting, and film-style instructions to suit your references, then generate. Choose an explicit aspect ratio when the final layout matters; auto follows an input image's aspect ratio. Resolution defaults to 4K. The supported targets are 1K, 2K, and 4K, with Gemini's native upscaler used for 2K and 4K. Image model selects between the two source-supported Gemini models. Seed -1 randomizes each run; a fixed seed requests similar responses but does not guarantee repeatable results, especially when the model or settings change. Generation instructions set the provider's general behavior; keep their default initially and use Prompt for creative direction. Provider response can request IMAGE or IMAGE+TEXT. Only generated images are delivered by this Pipe; provider text is not connected to an output. References and prompts are sent to Gemini through a paid provider call. The result is a generative interpretation: exact identity, clothing, product details, and placement are not guaranteed. Uploaded alpha masks are not passed to generation, so uploaded transparency is not preserved by this workflow.

Ovis-Image text to image

Generate one image or a batch from text with Ovis-Image. Describe your subject, setting, composition, lighting and style in the positive prompt; include any text you want rendered. Optionally describe unwanted elements in the negative prompt, or leave it empty. For a first run, replace the example positive prompt and keep width and height at 1024 pixels and batch size at 1. Width and height use multiples of 16; together they determine the aspect ratio. Batch size sets the number of returned images. Larger dimensions and batches require more memory. Advanced sampling defaults are 20 steps, prompt guidance (CFG) 5, Euler, the simple scheduler, denoise 1 and sampling shift 3. Steps set the denoising step count; CFG controls prompt guidance. Sampler selects the sampling algorithm, while scheduler distributes noise levels across steps. Sampling shift adjusts the model's noise schedule. Keep these defaults for the first run. Seed -1 selects fresh random noise each run; a fixed seed reuses the initial noise with otherwise unchanged settings. This tool generates from an empty latent and accepts no source image. Denoise ranges from 0 to 1; keep 1 for full text-to-image generation. Lower values shorten the noise trajectory and do not preserve an uploaded image. The output returns every image in the batch. The runtime needs the fixed source files ovis_image_bf16.safetensors in diffusion_models, ovis_2.5.safetensors in text_encoders and ae.safetensors in vae, plus the Sweet Tea output node pack. Model selection is fixed in this Pipe.

Kandinsky 5 image to video

Animate an uploaded still image into a short, silent video with Kandinsky 5 Lite and a scene and motion prompt. For a first run, upload your starting image, replace the example flower prompt with a description matching it, and describe the subject movement and camera movement you want. Keep 768 × 512 pixels, 121 frames, 24 FPS, and the default sampling settings. The default output is an H.264 MP4 lasting approximately five seconds. The source image is resized with Lanczos and center cropping to the selected dimensions. Choose an aspect ratio that keeps important subjects in view. Use positive dimensions of at least 16 pixels, divisible by 16. Uploaded transparency is not preserved: this path uses the image loader's IMAGE output, not its alpha mask. Frame count controls how many frames are generated; use counts of 4n + 1, such as 121. FPS controls how quickly those frames play and does not generate or interpolate additional frames. This Pipe uses the fixed five-second model; increasing frame count does not select the template's ten-second model. Use the optional negative prompt to describe unwanted content. Prompt guidance (CFG) controls conditioning strength. Seed -1 requests a fresh random seed each run; use a fixed seed when comparing settings. Steps, sampler, scheduler, sampling shift, and denoising fraction are advanced sampling controls. Retain their defaults initially. Denoising fraction is not a direct image-preservation slider in this graph. Keep generation batch size at 1 for a single clip. Larger batches feed the same video assembly path rather than separate clip outputs. Video output settings select the container and codec; retain MP4 and H.264 for the first run. No audio input is connected. Larger dimensions or more frames increase processing and memory requirements.

Stable Audio 3 audio generator

Produce MP3 audio from a text description using Stable Audio 3 Medium Base. No uploaded media is required. Describe the instruments or sound source, mood, texture, space, and desired progression; include BPM when it matters for music. For a first run, keep the supplied lo-fi description, Music category, 150-second Duration, prompt expansion enabled, and batch size 1, then generate. For an isolated hit, select One-shot and set a short Duration appropriate to its decay. Enter category values exactly as Music, Instrument, SFX, or One-shot. They select expansion instructions for the same audio generator, rather than separate generation models. Prompt expansion uses Qwen to elaborate your description. Duration sets the audio latent length independently and is converted to an integer for the expansion template. Use whole seconds to align these requests. The downstream duration input accepts 1–1000 seconds; this is an input range, not a guarantee of useful model behavior throughout that range. Category templates contain their own length suggestions, so rewritten length wording may differ from your requested Duration. Disable Expand prompt to send your description directly to audio conditioning. Category and expansion tuning then do not affect the selected prompt. Include desired tempo and length wording yourself. Expansion sampling off has a different purpose: expansion remains selected, but text generation uses no sampling. With sampling on, the expansion seed, temperature, and token selection controls affect the rewrite. Expansion token limit is measured in text tokens, not seconds. Thinking requests model-supported thinking behavior; the graph sends the resulting text directly to audio conditioning without a separate reasoning-text removal stage. Negative prompt supplies exclusion conditioning; leave it empty for the supplied default. Advanced Sampling controls the audio seed, steps, guidance, sampler, schedule, and denoising amount. Keep Denoise at the supplied 1 for full generation from an empty latent. Audio and expansion seeds are independent; -1 randomizes each run. To hold sampled expansion and audio noise choices fixed, set both seeds and retain the other settings. Audio batch controls how many duration-sized latents are sampled together. Longer clips, more steps, and larger batches increase generation work. Instrument and loop descriptions guide generation but do not guarantee isolated stems or seamless loop boundaries. MP3 quality selects V0 variable bitrate, 128k, or 320k compression and does not change the generated audio content. The source specifies stable_audio_3_medium_base.safetensors in checkpoints and t5gemma_b_b_ul2.safetensors in text_encoders from https://huggingface.co/Comfy-Org/stable-audio-3, plus qwen3.5_2b_bf16.safetensors in text_encoders from https://huggingface.co/Comfy-Org/Qwen3.5 for expansion.

Boring_e621 Negative Embedding: Enhance Images Stylistically AND Topically

This is a download site for the negative embedding boringe621. Its original huggingface repository, which goes into more detail about how it was trained and what it does, can be found here .

TextualInversionSD 1.4

FLUX.2 [klein] 4B Base1024x10248GB VRAM

Grok video

Generate one short video from a starting image and prompt, or from text alone with grok-imagine-video. For a first run, upload a starting image, keep Source image enabled, and replace the example prompt with a description of your scene's action, camera movement, and atmosphere. Keep grok-imagine-video-1.5, 720p, and the five-second duration, then run the Pipe to receive the video result. Grok-imagine-video-1.5 requires a starting image and supports 480p, 720p, and 1080p. For text-only generation, select grok-imagine-video and bypass Source image; choose 480p or 720p. That model also accepts a starting image when Source image remains enabled. Duration is measured in seconds, from 1 to 15. Aspect ratio sets the video's proportions; auto leaves the choice to the generator. Resolution sets the actual output resolution, regardless of resolution words such as 4K in the prompt. Leave Seed at -1 for a fresh randomized value each run. The seed controls whether the provider node re-runs; repeating a value does not reproduce the same video. Advanced Video output controls let you select the file type and codec; keep auto for the first run. Generation uses the Grok provider, sends the prompt and any starting image externally, and involves a provider-priced call.

Flux 2 Klein Base text to image

Create images from a text description using Flux 2 Klein Base 4B. This tool uses the base model with separate positive and negative prompts and adjustable CFG guidance. For a first run, replace the example positive prompt with your subject, setting, composition, and desired visual style. Leave the negative prompt empty and keep the source defaults: 1024 × 1024 pixels, one image, 20 steps, CFG 5, and Euler sampling. Run the Pipe to receive the generated image in Sweet Tea. No uploaded image is required. Width and height set both the latent image dimensions and the resolution used by the sampling schedule. Enter positive multiples of 16, at least 16 pixels. Together they determine the aspect ratio. Images per run creates a batch with the same prompt and dimensions; larger dimensions and batches require more memory. Use the negative prompt to describe elements or visual traits to discourage, or leave it empty. CFG adjusts the influence of positive and negative conditioning; at CFG 1, the negative prompt has no effect. In the advanced controls, steps set the sampling iteration count and sampler selects the sampling algorithm. Keep 20 steps and Euler for the source setup; changing either can change the result. More steps do not guarantee a better image. Seed -1 selects fresh random starting noise each run. Use a fixed seed to compare prompt or sampling changes with the same starting noise. This Pipe returns only the base-model branch's images. The separate distilled four-step Klein model belongs to the sibling tool and is not a selectable mode here. The diffusion model flux-2-klein-base-4b.safetensors, text encoder qwen_3_4b.safetensors, and VAE flux2-vae.safetensors are fixed dependencies. The output transport returns images to Sweet Tea without permanently saving them through ComfyUI.

ModeloCivitai

Atompunk Style

Follow me to make sure you see new poses and Nobodys when I post them. Things move fast on this site, it's easy to miss.

TextualInversionSDXL 1.0

Recompose a graphic for a new format

Recompose a flattened graphic for a new aspect ratio through generative background extension and layout rearrangement. For a first run, upload a graphic, keep Nano Banana 2 selected, and use its default 16:9 aspect ratio and 2K resolution. Read the layout instructions and adjust them to identify the text, logos, imagery and placement constraints you want preserved. Run the Pipe and inspect the returned graphic. Nano Banana 2 and Nano Banana Pro receive the same source image and layout instructions, but have independent aspect ratio, resolution and seed controls. Configure the settings for the model you select; the model choice determines which image reaches the output. Choose auto aspect ratio to match the source format. Nano Banana Pro offers 1K, 2K and 4K resolution settings. Nano Banana 2 also exposes thinking level, temperature and top p; its supplied resolution default is 2K, and the supplied schema does not enumerate other accepted values. A seed of -1 randomizes each run; fixed seeds provide best-effort repeatability without guaranteeing identical results. The provider response format can request IMAGE or IMAGE+TEXT. Only the selected model's final image is returned; text responses and thinking images are not included. This edits a flattened image rather than editable design layers. Text, logos, proportions and placement can change despite preservation instructions, so inspect the result before using it. Uploaded transparency is not preserved end to end: only the loader's image output is connected, not its alpha mask. Google Gemini API access is required; provider generation sends the source image and instructions externally and can incur provider charges.

1024x10248GB VRAM