Comfyui batch prompts not working. Find more, search less Explore.

Comfyui batch prompts not working I would love to know if there is any way to process a folder of images, with a list of pre-created prompt for each image? I am currently using webui for such things however ComfyUI has given me a lot of creative flexibility compared to what’s possible with webui, so I would like to know. Hopefully waiting for another build. I did not find any article on the internet related to this issue. select_every_nth will take every frame at 1, ever other frame at 2, every 3rd frame at 3 and so on if you need it to skip some. Manage code changes Discussions. md at main · Psynian/comfyui-dynamicprompts-batch Also, in A1111 I was switching between Models with Dynamic Prompts to get samples for thumbnails in the Model Helper extension. Unfortunately, due to how the batch process works in ComfyUI, Preview Chooser is kinda useless. Check out an example here: How to increase batch size and batch count in comfyui? I want to make 100% use of my GPU and I want to get 1000 images without stopping. I searched around but it doesn't seem like you have provided a solution to roll back the manager to a specific version I want, I tried Snapshot Manager, but it did not work as well. Combinatorial prompt: A {red|green|blue} ball by me When batch size is set to 1, only 1 image is returned. You signed in with another tab or window. It outputs result, a STRING, which is (initially) the first line from lines with the format applied. A new Image2Image function: choose an existing Welcome to the unofficial ComfyUI subreddit. V1 Desktop Feedback. The regular prompt schedule works on everything with a float or int like you've observed. For example, if I use the prompt Oil painting portrait of [Joe Biden: Donald Trump: 0. I am not sure if it is the latent that is changed on each bath, if it is the seed that is scrambled or both. Then pull the conditioning outputs to the sampler. So I deleted the entire thing and started over. I've tried a few techniques (I'm talking comfyui, I've never tried A1111) and BREAK is not one that works for me. I ended up building a custom node that is very custom for the exact workflow I was trying to make, but it isn't good for general use. The syntax is [keyword1 : keyword2: factor] factor controls at which step keyword1 is switched to keyword2. I thought I had renamed it correctly from my. Any I managed to get it working, I just copied the example workflow from github and added onto it. bat file to accomplish things like automatically bringing up the GUI, etc. queue prompt not working even after everything is correct. png”, “table. The node I used was the Value I have the same problem. In 1111 using image to image, you can batch load all frames of a video, batch load control net images, or Can I create images automatically from a whole list of prompts in ComfyUI? (like one can in automatic1111) Maybe someone even has a workflow to share which accomplishes this, just like it's possible in automatic1111 I need to create images from a whole list of prompts I enter in a text box or are saved in a file. b) Open a command prompt and navigate to the folder containing get-pip. Let me explain. The correct way to define the batch size is in the empty latent image node where you set the resolution. But cached does not mean files are re-written by re-running the same API call. Look for a comfy example work flow for sdxl Hi thank you for your answer. Experimented with different batches, prompts, models, etc, but to no avail Any ideas what could be stopping my animation? Here's a warning I get: *"*WARNING: Sequential prompt keyframes 3:150 and 4:10 are not monotonously increasing; skipping interpolation. npm install brings back "'npm' is not recognized as an internal or external command, operable program or batch file. The proper term is prompt scheduling. Whereas traditional frameworks like React and Vue do the bulk of their work in the browser, Svelte shifts that work into a compile step that happens when you build your app. done. Once you build this you can choose an output from it using static seeds to get specific images or you can split up larger batches of images to reduce ram usage and stop running out of ram errors when you are working with batches of larger images. I use a simple prompt of a black dress, I get a woman in a white bra and panties. it could be just a better understanding on my part. Open 3 tasks done. r/StableDiffusionInfo. Class name: ImageBatch Category: image Output node: False The ImageBatch node is designed for combining two images into a single batch. Question - Help Start at index 0 to load the first image in the folder and set your batch count (under Queue Prompt --> Extra Options) to the number of images in the folder. The If I drag a workflow onto the screen, it stays empty. However, there is not a way to Heads up: Batch Prompt Schedule does not work with the python API templates provided by ComfyUI github. Please check the logs for more details. Get the same frame all over. This really feels like more of an alpha build than a beta. It works, but kinda silly way to make ComfyUI run over night. Experimented with different batches, prompts, models, etc, but to no avail Any ideas what could be stopping my animation? Here's a warning I get: Welcome to the unofficial ComfyUI subreddit. Then I went to ComfyUI & began installing modules. For T2I, you can set the batch_size through the Empty Latent Image, while for I2I, you can use the Repeat Latent Batch to expand the same latent to a batch size specified by amount. 5520x4296 upvotes I successfully rolled back Comfyui itself, but it seems like the problem is not from them as the batch count feature still works extremely slow. I’m trying to create a simple loop that pulls a prompt sequentially from a list as it it Simply running the executing the prompt multiple times, either by smashing the "Queue Prompt" button multiple times in comfyUI, or changing the "Batch count" in the "extra options" under the button. Consistent character using only prompts - works across checkpoints and LORAs 10. json or workflow_example. Presumably, you would have a latent node batch_size of 1 here. bat if you are using AMD cards), open it with notepad Reply reply AstronomerPutrid9214 • i have tried this but still its not working Reply I just moved my ComfyUI machine to my IoT Hello, with same input json, comfyui generate different images every time. The nodes provided in this library are: Follow the steps below to install the ComfyUI-DynamicPrompts Library. The batch count is under extra options in the menu. This is NO place to show-off ai art unless it's a highly educational Welcome to the unofficial ComfyUI subreddit. It's bleeding through sometimes, but not consistently. I think the intended workflow here is to just press several times on the Queue Prompt button. I have objects in a folder named like this: “chair. Sometimes when you click the Queue Prompt, nothing happens and it seems the “Queue Prompt” is not working. Click the This tool is perfect for artists who want to explore new ideas without the hassle of manually crafting detailed prompts. I did Install Missing Custom Nodes, Update All, and etc etc, but there are many issues every time I load the workflows, and it looks pretty complicated to solve it. For all I can tell it completely ignores the X and Y inputs now and only refers to the latent batch size. My comfyui for some reason became unstable and what ever I typed in my prompt, say a castle, i would get a nsfw image of a woman. Batch Images Documentation. can yo This video will show you a super easy trick to batch process your text prompts in ComfyUI and create a whole Tired of making one image at a time in ComfyUI? This video will show you a super Welcome to the unofficial ComfyUI subreddit. And I mean batch on the right. The special sauce though being instead of just rendering fresh we are I came up with a way using a custom node. You switched accounts on another tab or window. Nothing in the prompt was followed. You signed out in another tab or window. like the thing that has save load, refresh, where you view history. Unfortunately, the uninstaller seems to do almost nothing, other than removing the shortcut from my desktop. Eah time I need to restart comfyui V1 on Windows 11, nothing happens. It's wise to load and process those images one by one in a batch task like automatic1111. Maybe I will now. ba When opening comfyui a green circle spins eternally and does not open. Since I used ComfyUI, I downloaded tons of workflows, but only around 10% of them work. We demonstrate how you can create a short video depicting the change of seasons. Open jasoncow007 opened this issue Dec 10, 2024 · 1 Credit should go to Dennis for verifying that my. It is concerning me. -> The image preview in the new ComfyUI frontend does not work question Further Welcome to the unofficial ComfyUI subreddit. To use this custom node (located within the 'utils' submenu), simply connect your positive prompt to it, which will then output the joined prompt. I update from update folder with update_comfyui. compatable with a/framesync and a/keyframe-string-generator for audio synced animations in Comfyui. It provides nodes that enable the use of Dynamic Prompts in your ComfyUI. but don't spam all your work. 0 seconds: F:\AI 2024\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\sdxl_prompt_styler Expedition not working upvotes r/StableDiffusionInfo. bat file (or to the run_cpu. And above all, BE NICE. Notifications You must be signed in to change notification could it be possible work with batch prompt schedule #5. These commands The Inspire Pack has a LoadImagesFromDir node: you write the path of the folder containing the images, use that node as an input, and the workflow will work on every image, making one output per image. SilverPrinciple4998. bat but the problem was that it was actually named my. Also, if this is new and exciting to you, feel free to post, but don't spam all your work. The was-node-suite-comfyui has the custom node you need to load images from a folder, and you can use Auto Queue to iterate the image loading. By applying both a prompt to improve detail and to increase resolution (indicating as percentage, for example Path of SD(Portable) and extension ComfyUI is D:\stable-diffusion-portable-main\extensions\sd-webui-comfyui - its not working Path of ComfyUI(standalone portable) D:\ComfyUI_windows_portable\ComfyUI - its working fine, but not extension If i change standart "ComfyUI install location" to D:\ComfyUI_windows_portable\ or D:\ComfyUI_windows Welcome to the unofficial ComfyUI subreddit. (see Installing ComfyUI above). Scheduled prompts, scheduled float/int values and wave function nodes for animations and utility. ComfyUI-DynamicPrompts is a custom nodes library that integrates into your existing ComfyUI Library. Put the path (has to be on local machine running where the comfyui server is installed). Welcome to the unofficial ComfyUI subreddit. For such a small task, it may seem so, but implementing such a batch process is not so easy to do when you consider larger workflows. 5-Turbo. Is ther a key command to get it back? Thoughts? You signed in with another tab or window. Is this achievable? ComfyUI node version of the SD Prompt Reader - Issues · receyuki/comfyui-prompt-reader-node Filename not output for each image from SD Prompt Reader when using SD Batch Loader node bug Something isn't [BUG] - Use new menu and workflow management. bat. Sounds easy but when i set a batch count, no matter what i do the images are different. Currently the process works off a batch of conditionings, and schedules for settings like denoise, and loops through the prompts. mp4 The Ksampler simply can't handle a batch of values from the batch value schedule which is the posted issue. py", line For the newest comfyui I am not sure. #58 opened Jun 26, 2024 by AlexBenjarmin. As long as you use I seem to have problem with "connecting the prompt" with the video reference (depth controlnet and ipadapter - xl model), I managed the batch prompt to work without the ip adapter and controlnet: Video_00023. Closed photogsy opened this issue Oct 25, 2024 · 5 comments Closed Dynamic_Prompt_Working. To get the workflow as I've modified the encode method in ComfyUI/nodes. Ideally, I'd love to leverage the prompt loaded from the image metadata (optional), but more crucially, I'm seeking guidance on how to efficiently batch load images from a folder for subsequent upscaling. Hey there! I have a super newbie question: is it possible to use the same prompt (both positive and negative) for two samplers that have different checkpoints? You should be able to connect multiple samplers to the same prompt. Here's a The equivalent of "batch size" can be configured in different ways depending on the task. Either way, just search for comfyui wildcards nodes. I think Was has wildcard nodes. Extremely slow import #55 this is also happening to me after updating comfyui, even the default workflow is not working! Simply saying it doesn't work doesn't allow us to understand anything. Additionally, batches must have the same size due to a constraint. Want 10 images I don't know about settings, but there are nodes for making dynamic prompts with wildcards. You don't necessarily need a PC to be a member of the PCMR. Lists allow for any possible output, whereas batches are constrained by images, latents and masks. There is nothing in the console. A new Face Swapper function. Got an issue here when I try to install your node. Your question I use batch count to generate 100 images at a time, using multiple different tabs for different generation before I go to bed. i can't find out how to get it back. Lilly has them, but something happened to my comfy when I installed Lilly, so I had to reinstall comfy. Why is this "else" and "return i wish use batch prompt scheduler to produce same element for different image,such as a car,then inpaint different images' car . Luckily I found the simplest solution: Just link the Loadcheckpoint Node to Batch Prompt Schedule (Fizznodes), Sometimes when you click the Queue Prompt, nothing happens and it seems the “Queue Prompt” is not working. I've seen a few YT videos talk about using a specific . It sounds like what you're doing is #1, but you're expecting behavior in #2. Please keep posted images SFW. This is great for prompts so you don't have to manually change the prompt in every field (for upscalers for example). My workflow is as simple as that: "load image from list/dir (inspire pack) -> upscale by model (+ load upscale model) -> save image (here: batch)" Problem: re The new Comfy app is just in an unusable state for me at the moment. ) is the batch, so you can run multiple at the same time with different settings. the prompt shouldn't be revisited during a batch. Learn more by watching @poisenbery's vids on There’s already some good advice here, but if you can program in Python, consider looking into how the API works. I think I have a reasonable workflow, that allows you to test your prompts and settings and then "flip a switch", put in the image numbers you want to upscale and rerun the workflow. #4400. , but I'm not finding it as expected in the base folder of comfyUI, and maybe I just didn't see it in the readme. Oh that's because ComfyUI has an optimization where it only Welcome to the unofficial ComfyUI subreddit. The generated graph is often exactly equivalent to a manually built workflow using native ComfyUI nodes. "Smart" Search and Sort functions (Search works by both name and theme, so you can search for things like "cute") Easily add your own images to the Custom category; Automatic metadata extraction for prompt tags on custom images (only works with default ComfyUI nodes ATM, works well with FLUX though!) I can't load any workflow or manager. If I click the Queue Prompt button, the workflow becomes visible, but it's still unusable. However, after a recent update, I noticed the "got prompt" process has become extremely slow. It's like my ComfyUI screen is just dead. Even better than the JSON workflow files, images produced by ComfyUI have that JSON info embedded within each image. BarbaRoxaRules opened this issue Dec 3, 2024 · 1 comment Open Traceback (most recent call last): File "G:\ComfyUI_windows_portable 2\python_embeded\Lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process. the diagram doesn't load into comfyui so I can't test it out. I can't zoom in or out. I don't think it works as an input for a KSampler though I stand corrected, it works, I just had a problem in my workflow x). ComfyUI custom nodes for Dynamic Prompts with Batch Functionality - comfyui-dynamicprompts-batch/README. This is probably the I've been experimenting with batch generation and works fine with the Image Batch and Mask Batch of the WAS_Node_Suite. And I'm not sure where the proper list of ComfyUI commandline args would be. Discuss all things about StableDiffusion here. a search of the subreddit Didn't turn up any answers to my question. You can open it from the Help menu. I want to load it into comfyui, push a button, and come back in several hours to a hard drive full of images. It now has a "text"input, but nothing can be plugged into it I am new to ComfyUI and I am already in love with it. My problem is that the auto queue button under extras stops working after 100 images, if I am not using the computer (it is unattended). Default Prompt Not Working, Help . Note that I started using Stable Diffusion with Automatic1111 so all of my lora files are stored within StableDiffusion\models\Lora and not under ComfyUI. One of the reasons is that there is a "caching" of ComfyUI, it will NOT queue I have an SDXL checkpoint, video input + depth map controlnet and everything set to XL models but for some reason Batch Prompt Schedule is not working, it seems as its only AnimateDiff, a custom node for Stable Diffusion within ComfyUI, enables the creation of coherent animations from text or video inputs. Switched to BastardAIImpossible when A1111 crashed. All reactions It will swap images each run going through the list of images found in the folder. 1> I can load any lora for this prompt. Authored by FizzleDorf Welcome to the Official Toon Boom Harmony subreddit. There are two dynamic prompt extensions. But basically use it same as load image node. I copy/pasted from different workflows I found. Maybe i have not done some things right but installing a software shouldn’t be such a problem. py. Tried clearing it out and changing prompt["12"]["inputs"]["choose file to upload"] to path also, but no dice. Extra options checkbox below "queue prompt" then you can tick the checkbox beside the slider to make it auto queue. All features Where is the queue prompt button in the updated ComfyUI? Logs. Single image works by just selecting the index of the image. png) onto ComfyUI. Here is a part of log found in “main. Collaborate outside of code Code Search. The Batch Prompt Schedule in ComfyUI is a powerful When you click “queue prompt” in ComfyUI, it actually sends a POST request with the whole workflow as JSON data to http://127. Posts about hiring will be Comfyui batch prompts not working You signed in with another tab or window. However, there is not a way to handle batches of text inputs. Please upload the full log. Welcome to ComfyUI Studio! In this tutorial, we're showcasing the 'Default Prompt Batch' workflow from our Ultimate Portrait Workflow Pack. Hello, BatchPromptSchedule in Comfy UI is only running the first prompt, I had it working previously and now when running a json that did go through the scheduled prompts it will only use the first. I tried some things i read here around but they did not work. To use it, you fill lines with many lines of input and you use If can also be a fixed seed in the sampler, if you have a fixed seed and don't modify the prompt or any other parameter, after 1 queue comfyui won't generate the image, because it would be exactly the same as the one already rendered before. txt! Fixing that was key to this problem. This workflow is Batch-processing images by folder on ComfyUI . Dynamic prompts. I believe it's due to the syntax within the scheduler node breaking the syntax of the overall prompt JSON load. After borrowing many ideas, and learning ComfyUI. why are all those not in the prompt too? It was dumb idea to begin with. batch works correctly in comfy ui. FYI, you can skip the prompt nodes when using the Efficient Loader and put the prompts in that node. Right now I am using CR Integer List to create batch generations from a single ksampler When I pipe the batch of images into a Paste By Mask node with randomized mask locations, it creates 3 different versions of the base image with 1 sticker applied to each version—- Using the Choosing Multiple Values function with dynamic prompts in ComfyUI does not work. The text was updated successfully, but these errors were encountered: One would assume that it automatically interpolates smoothly from one prompt to the other. Please share your tips, tricks, and workflows for using this software to create your AI art. A new Prompt Enricher function, able to improve your prompt with the help of GPT-4 or GPT-3. I was trying to research how the negative prompt works, but the only thing I could find were reddit posts and youtube videos that said the same thing In this video, we show the transition between several prompts. I don't understand myself, tbh. It’s pretty easy to make a workflow, save it as an api version, and then write a script to import your workflow, change small parts (change what image is loaded, change the output folder, randomise the prompt) and then submit it. issue I use Load Images [Deprecated], index load cap 0, start index 0, then an Ultimate Upscale, then an Image Save. Diffusers allows this by passing multiple prompts to the pipe, which are stacked in a tensor and pushed through the pipeline. My ksampler is working fine, but after the update, the real-time preview is no longer functioning. if you are using this node please make sure Prompt control has been almost completely rewritten. If you give me a workflow, I can cook something up quickly. AnimateDiff, a I've been trying to do something similar to your workflow and ran into the same kinds of problems. It's similar to the concept of inheritance. When batch size is set to 3, 3 images of the same prompt is returned. png”, “door. log” file in comfyui V1 folder Hi newbie here made this batch face restore workflow. Cheers! 🍻, and thanks again, for the help this thread has provided me, Thanks + understood. It is Beta of course. They work exactly the same as the corresponding terms in A1111/SD. My prompt is not producing anything that he is making. I Look for batch_size in the Empty Latent Image node. Next to your “manager”. ADMIN MOD Looping batch prompt and batch images in sync . 1: 88: November 28, 2024 ComfyuI Impact pack not installing. The only workaround I've found thus far is make an auto clicker script that will click every 30minutes or so. I'm trying to build an iterator to process a batch of images on a folder. A portion of the Control Panel What’s new in 5. mp4. Instead of using techniques like virtual DOM diffing, Svelte writes code that surgically updates the DOM when the state of your app changes. I try to close the window I try de “Quite Comfy” from tray icon bar The only way I found is to use Windows Task manager to kill the process to be able to switch it off then launch it manually again. Best I could figure out was using the wasnode batch loader on increment and setting the The batch image node takes single images and makes them into a batch. I am not near computer. I'm not sure if it's due to a conflict with custom nodes or changes in the settings, but before the update, the real-time preview feature was working normally. As I wrote recently on this sub, in terms of gaining separation between two prompt items, the following is what I've found: Punctuation (comma or full stop) - yes new line - no cur_prompt_series[i] = prepare_batch_prompt(cur_prompt_series[i], max_frames, i, prompt_weight_1[i], The text was updated successfully, but these errors were encountered: All reactions Say you have a folder of images of poses you want to mimic using the ControlNet OpenPose model in ComfyUI. Notifications You must be signed in to change combinatorial It's not so simple. It now uses ComfyUI's lazy execution to build graphs from the text prompt at runtime. [Help]: Create filename templates: Not really working I'm new to ComfyUI and in the process to adapt my usual workflow and behavior to this new program. The qeue runs, but does not produce an XY grid and not enough images. I want to generate same image so i add control_after_generate in Ksampler, but it does not take any value as input. png”, etc. Thanks! All reactions. Wasn't able to get it working just using the nodes. py to a folder on your computer. It is a number between 0 and 1. I've submitted a Perhaps next time you could post your question in the issues rather than in the discussions, I'm not sure why I don't receive notifications for discussions. Topics Trending Collections with the latest update it has stopped working. I like his videos and results so I was trying to copy his work flow and use my wife's face to swap in. . and doesn't work correctly in a 11:11. loras in the prompt was dumb and Topics Trending Collections Enterprise Enterprise platform Install bat not working #498. i'm not sure why they seem to be working now. a) Download get-pip. as far as I'm concerned combinatorial is broken because of alphabetizes before it randomizes so it's pseudo random edit: Just tried Dynamic prompts nodes* but it doesn't seem to work. IMO. This needs to be corrected to delete the entire folder structure from my PC. Works great if you want to queue [1,100] images. i can use a controlnet between batch prompt scheduler and the guilder, is the batch prompt scheduler working Plan and track work Code Review. next install pip using. Is it possible to get a new download link? Related topics Topic Replies Views Activity; Can't go to desktop. I can not get Comfyui working. Has anyone been able to get Dynamic Prompts working since last week? Whatever was done in early October really messed up Dynamic Prompts. If I try to add a node, the node list pops up, but no node pops in when I select it. Once I've amassed a collection of noteworthy images, my plan is to compile them into a folder and execute a 2x upscale in a batch. The text was updated successfully, but these errors were encountered: 8:33 [ComfyUI] Adjusting the positive and negative prompts 9:01 [ComfyUI] Test generating (ControlNet working; Dynamic Prompts not working) 9:15 [ComfyUI] For Dynamic Prompts, setting batch SIZE to more than 1 is of no use, since the same seed is used for all the wildcards 9:29 [ComfyUI] For Dynamic Prompts, setting batch COUNT to more than 1 Welcome to the unofficial ComfyUI subreddit. If I try to drag the screen around, it doesn't move. Sign in Product ShmuelRonen / ComfyUI-AstralAnimator Public. Not a stupid question because it is very confusing if you've come from Auto1111 to understand how seeds work in ComfyUI. Navigation Menu Toggle navigation. And oddest thing: I've used the xy plot script from "Efficiency Nodes for ComfyUI", but ever since my first installation attempt of Manager, it doesn't work anymore. “‘wmic’ is not recognized as an internal or external command, operable program or batch file. The whole diffusers library operates under the assumption that the first dimension of the various tensors (prompt, negative prompt, seeds, etc. One alternate method is to use the api and a python script. Green is your positive Prompt. I haven't touched ComfyUI in a few months so I am just tinkering with it now. I have a text file full of prompts. It's not about the hardware in your rig, but the software in your heart! Join us in celebrating and promoting tech, knowledge, and the best gaming, study, and work platform there exists. Find more, search less Explore. ) I call mine FormattedLineByIndex, and as inputs it takes a fmt, a STRING, and lines, a multiline STRING. However, I'm pretty sure I don't need to use the Lora loaders at all since it appears that by putting <lora:[name of file without extension]:1. This is not a forum for recruiting animators. adieyal / comfyui-dynamicprompts Public. And for each of the objects, I want to prompt with different style like this: Is it possible to iterate a ComfyUI workflow over a batch of video clips within a folder for a vid2vid workflow? I'm trying to remaster a CGI video with realism, but it's got about 90 clips I need to process, and I don't want to have to hold the whole video in memory to process all 7 minutes. Next. Sorry should have uploaded troubleshooting data. Due to the random nature of the prompts, not every generated image will be perfect The extra options from the control panel, from what I can see, have a batch count (no batch size) option; the only thing the option does, I think, is queueing up a number of batches of size 1 one after the other (basically the same thing as if I Check if you have python installed properly. I'm trying to get dynamic prompts to work with comfyui but the random prompt string won't link with clip text encoder as is indicated on the diagram I have here from the Github page. No response. Is your ComfyUI and all custom nodes updated? If so, the issue could be caused by a custom After installing and opening ComfyUI, I get the following error: Was not able to start ComfyUI. ) I'm also facing the same issue after updating the ComfyUI yesterday. 1:8188/prompt . Is this to say that negative prompts don't work? Not at all! But, I don't think they work in the way that most people believe. Reload to refresh your session. ” How do i run this? Related topics Topic Replies Views Activity Welcome to the unofficial ComfyUI subreddit. I am doing this now. However, the other day I accidentally discovered this: comfyui-job-iterator (ali1234/comfyui-job-iterator: A for loop for ComfyUI (github. By specifying the starting index and the length of the subset, you can efficiently manage and manipulate your image data without having to handle the entire batch at once. sharing to see if it helps out anybody and please let me Welcome to the unofficial ComfyUI subreddit. com)) . It generates all images within the batch with the same prompt even though I'm using this: Red {toy train|toy car} In A1111 Dynamic prompts extension by the same author, Welcome to the unofficial ComfyUI subreddit. py, I've been experimenting with batch generation and works fine with the Image Batch and Mask Batch of the WAS_Node_Suite. I've Use batch_count instead of batch_size. Set batch to run the number of images you have (if 10 images in folder set batch to 10). Running the combinatorial prompt example json workflow. 0. Simple iterate through a directory to delete matched files not working upvote to the run_nvidia_gpu. I've modified the encode method in I also had the same issue with blending images using Batch Prompt Schedule. I think it works like a charm but maybe can get better. 5] for 30 sampling steps. Skip to content. it should be revisited and randomized between each batch Unless it's combinatorial. It'd pick a random selection for the very first generation then all pictures generated as part of that batch are the same. They will then behave the same as if you have generated a batch of images using a ksampler. I liked the ability in MJ, to choose an image from the batch and upscale just that image. (If you restored using the snapshot feature, this phenomenon may have occurred as it forcibly adjusted to that version of ComfyUI. Closed lunatico67 opened this issue Aug 16, 2024 I received the download link a few days ago and wanted to download today but the link is not working anymore. I know one stopped working because it was replaced by the other. As such, I decided to uninstall the app. is GONE. sharing a comfyui workflow is super simple: drag and drop an image generated by comfyui into your comfyui window: boom. I can stil ctrl+enter to run it but would like to make a batch run. But when I skip_first_images will allow you to skip so many frames at the beginning of a batch if you needed to. Travel prompt not working. For animators of all skillsets who work with Toon Boom Harmony to share advice, get tips, and talk shop. install. So, it works for one picture then stops. 0. See comments made yesterday about this: #54 (comment) I did want it to be totally My understanding is that with "batch count" you can specify how many prompts and with "batch size" it's how many simultaneously, but I've often wondered: what if you want to run four different prompts at once with batch size?Can it be done? There was a bug in Dynamic Prompts that did this under AUTOMATIC1111 but I'd like to do it deliberately, in ComfyUI if possible. But it always stops working correctly after a So when I render batches, comfyui is processing them simultaneously, previewing just the first image. There are no more weird sampling hooks that could cause This issue appears to have occurred because you were not on the main branch. News & discussion on Data Engineering topics, including but not limited to: data pipelines, databases, data formats, storage, data modeling, data governance, cleansing, NoSQL, distributed systems, streaming, batch, Big Data, and workflow This is particularly useful when working with large datasets or when you need to isolate and process a specific segment of your image batch. The last two lines of the log file are like this: the value on prompt["12"]["inputs"]["image"] doesnt work as a full path. How One Button Prompt Works. " and trying to install npm with pip install npm brings back "E:\Comfy ComfyUI Academy - a series of courses designed to help you master ComfyUI and build your own workflows Thanks @AIWarper, Amazing work! Kendomland - land of the free and the men upvotes Took forever and might have made some simple misstep somewhere, like not unchecking the 'nightmare fuel' checkbox. ComfyUI now supporting SD3 upvotes eg: batch index 2, Length 2 would send image number 3 and 4 to preview img in this example. It's not the fault of the Preview Chooser developer/s, of course, it's just how it All my frames look exactly the same. It’s fixed now I just updated comfy again and updated all nodes and it started working. Please help. But its somehow "struggling" with IPAdapter and Depth map controlnet - as the batch prompt schedule does not apply: Video_00022. I'm using the default setup and The easy way would be a batch or list loader that will output the filenames, which I thought I had but the comfyroll one would throw errors when I fed it diff res images. ,,,so that the change could be directed by the prompt. Reply reply alecubudulecu • I was gonna ask about Welcome to the unofficial ComfyUI subreddit. 1. Random / branching prompt not working #5364. About the batch loader, are you able to use it now? If not, you can post screenshots and the workflow to the issues, so I can determine where the problem lies. This works but I want (-->>memory issues) to save this 1 image a time! When I start with a batch of 10, then Comfyui does his DaVinci Resolve is an industry-standard tool for post-production, including video editing, visual effects, color correction, and sound design, all in a single application! All creators, hobbyists to professionals, are welcome here. It's quite simple. Reply reply GalaxyTimeMachine • You'll need to copy any wildcard prompt lists from the old to new wildcards directory. I didn't even go back to the nodes file. bat should work but as he surmised, it was not named that. txt to my. Wasn't able to loop through all options in a single "Queue Prompt" like i hoped, but if I hit Queue prompts three times, it would go through the list of options. What I'm struggling right now is the problem that I can't name my saved files to what I want to (and used to). (The basic trick could easily be applied to your own node. You can also load the example workflow by dragging the workflow file (workflow_example. If the dimensions of the images do not match, it automatically rescales the second image to match the first one’s dimensions before combining them. One of the reasons is that there is a "caching" of ComfyUI, it will NOT queue the prompt if nothing changed in the workflow from the previous “Queue Prompt”. If I restart everything, it'll sometimes work normally for a few batches and sometimes won't. three times So my though is that you set the batch count to 3 for example and then you use a node that changes the weight for the lora on each bath. That means the prompt in steps 1 to 15 is Is there a node that allows processing of list of prompts or text files containing one prompt per line list or better still - a node that would allow processing of parameter sets in csv or similar spreadsheet format, one parameter set per row, so I can design 100K worth of prompts in Excel and let ComfyUI render them weeks at a time? Install this extension via the ComfyUI Manager by searching for One Button Prompt. bat, update_comfyui_and_python_dependencies. Other. But actually it just decreases one prompt until its strength reaches 0 and then uses the next I've been googling around for a couple hours and I haven't found a great solution for this. It seems like there are two competing things at play here: For example, having a wildcard node generate different prompts in the same batch? I know the {} syntax exists for front-end So my Queue prompt manager thing is missing. You just have to love PCs. random {red|green|blue} not working #57 opened Jun 25, 2024 by dangerweenie. odvgtd ovo rpt jfqda ufngqf rcheg tcy xvv icjwby gzsy