#opal-general・✨
1 messages · Page 3 of 1
is opal can be consumed like using an API request?
https://opal.google/_app/?flow=drive:/1BTmgilPbal0Q-c3lDJb60AIjrnX535vM&shared&mode=app
I have created this app so called billsplitter, this will being used when a group is dining in one table, but wanted to split the amount to pay per person, it has done as I expected
I think what made it slow because everytime the request is made, the interface is always different, so the agent making the interface again-n-again
and this is why I asked about using like API, so I can just make an API request with params of image and other required, then I can have response of JSON
@lost sapphire
I did the same suggestion, ZIOCAN (I was
Why not be super specific with the output prompt?
https://opal.google/_app/?flow=drive:/1Td8P6LYuymSjTwwFsUWzW5uXEGv_hJ78&shared&mode=app
@lost sapphire
Feature Request - Need an option to order the user inputs.
Don't know how the current order is defined (may be based on order of creation of user_inputs).
how to use opalhere ?
can you suggest me or tell me an example of the spesific terms? seriously interested here
Like in The Output Block. Describe all the Aesthetics, Color Palette & Formatting. Like A Number with Bold and has 25pt Font Size in Comic Sans Font. A Table with each persons name and the split amount on the otherside with a dollar/currency formatting with Italics for the Person Name with Font Size 16pt in Times New Roman (#F17604 for the Table Headers Fill & A Off White for the Table Header Text) (All Text should have an off-black color (put in Hexcodes for all colors))
Is what i am talking about
oh it will make the output not thinking too long, is that what the goals you're talking about ya?
Yep, The More Strict You are with your criteria the better for your case.
hi, how is start?
Fine
Hey all, is it possible to connect to external triggers. Example user input can be taken from whatsapp or external apis
Hey all, is it possible to connect to
today there is the update?
Why? Did an Update came?
Unable to use Opal on my iOS devices for 2 weeks.
Every time I try to login to Opal, I get the following loading screen that doesn’t progress even after 15 minutes. I have the same experience on my iPad and my iPhone 13 Pro.
Chrome version: 142.0.7444.148 (Official Build) stable (64-bit)
iOS 18.6.2 (22G100)
It works fine from Chrome on my Windows 11 PCs.
This loading screen occurs after I confirm Opal’s access to my Google account.
Unable to use Opal on my iOS devices for
Why? Did an Update came?
@grizzled girder is today the update on Opal?
is there a reference doc around supported file formats? Details
Unsupported MIME type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Can anyone actually access this thing? No luck on mobile
Can anyone actually access this thing?
Guys, I am new here. I just signed up for it's on the same loading screen from yesterday.
Hi! May I ask when Gemini 3 Pro and NanoBanana Pro are expected to be available in Opal? Also, if I’m on a Pro or Ultra plan, will NanoBanana Pro offer higher image-generation limits? I’ve noticed the current limits feel a bit restrictive. Thank you
Hello, since opal is out, im waiting to try it but i cant because my country isnt supported 🇨🇭 😥
do you know when it will be suported in switzerland?
also im curious to know why most of the things in google lab arent available in switzerland , does someone has an answer?
EU Regulations Primarily (which is quite important)
with a simple research you will find out how to access even in EU etc
i know with a vpn i can acces it, but im too lazy to use on😅
everything from the lab I am Trying to test is not available for my country (Germany) I would be interested in testing Opal, is working via VPN or is it tight to the account
Ah, I read the post before at the moment, so VPN is the goal
It would be so cool if Opal is integrated directly within Gemini!
Like how?
Listening
if I could ask gemini to understand the app structure and code a deployable app using gemini canvas?
it would be so cool!
Ohh, Gemini Auto creating an Opal App via the canvas?
Yea, That can be interesting
yashjit, how long have you used Opal for?
yess!
I've been using it since the day it was announced!
If not consistently, I still use it a lot to prototype my apps
Interesting, What have you found during your time on Opal?
I see. I don't use it consistently as well. I got a few applications particularly with Google Sheets. So, Most of the Time i would spend the time setting up the Sheet's Formulas (like an Search Function)
I really love to see a pre built structure of the app on opal connecting various gemini models like an arrow diagram. It makes it really easy to understand the workflow
Oww nicee to learn about your usecase 🙂
Like more in the area of how do you structure your prompts? I mainly stick to base model
Generating and Formatting a monthly report or an transactions history based off an Monthly Expense Tracker so on...
♊ 🍌 Gemini 3 and Nano Banana Pro should be available now
no deep research with gemini 3 pro??
no deep research with gemini 3 pro??
??? gemini 3 pro 😿
Greetings, Does Colab Notes work on Opal? Can i tell to execute a function which i have written in Colab notes?
Via an Asset Ofcourse
Hi! When using NanoBanana Pro on Opal. It feels quite limited. Is this an intentional rate limit or a temporary restriction? Thanks
Hi everyone, I have this problem when accessing everyone's Opal link via smartphone. Why ?? @grizzled girder
an error when using Gemini 3
:No text generated. Opal has limited quota and it was exceeded. Please try again later.
I really keep it very short and concise highlighting exactly what I need from Opal. It digs through the perfect required models and then make the perfect workflow
I think putting this simple prompt might work as well!
error gemini 3
Gemini 3 and Nano Banana Pro quota is extremely low right now! We're a bit tight on quota, so just wanted to let you all know that we're aware. We only released it silently and notified you all on discord, we're not publicizing these models being available otherwise since we know they're often overloaded. Bear with us and feel free to try these, but for any regular more reliable use, please use Gemini 2.5 and the first Nano Banana model 🙂
Respectfully I have a question. When will opal be available in Poland!?
We're still waiting on legal counsel to help us unlock Poland access, so sadly no ETA yet!
Btw, I have heard some info about an Looper Function from you, a couple of weeks ago? Is it scrapped or is it still in progress (it is quite a hard feature to pull off since there are some variables such as how should the process happen and such in the background or should it make the current output inaccessible etc.) Just Curious to know about it after some time. Thank You - Revamp
My never ending quest to get opal to login has hit a new type of error
Hi, I am trying to connect from Spain, and it is saying that it is available in my region. Is this correct?
Hi, can I ask is there any ETA on when Opal will be available in the UK?
APP test: https://opal.google/?flow=drive:/1QDLh7IuFNLVEBXiHktbNYnELjz0JW2BW&shared&mode=app
The APP will crash when uploading a long video because it is currently processing the video as base64. Meanwhile, we already have the API at https://generativelanguage.googleapis.com/upload/v1beta/files, but it isn’t being used. Could you update it so the APP can handle and analyze longer videos?
APP test: https://opal.google/?flow=
opal is so cool
will it be integrated to Gemini Apps like Canvas and Nano Banana or will have its own app/website
Did I read something about an Opal integration into something? Was I imaginging things? 🤣
Imaginging? Hello extra g
My never ending quest to get opal to
Did I read something about an Opal
will it be integrated to Gemini Apps
any way to import from aistudio to opal
Hi, can I ask is there any ETA on when
Hi, I am trying to connect from Spain,
Btw, I have heard some info about an
Thank you!
No Problem
I do have an Privacy Related Question.
Lets say i prompt a generate block to search the web. (As far as i am aware, It searches the web on the account which activated the prompt's behalf)
Does this data getting stored or used by Google in any form or manner?
Lets say it enters an website and (it accepts cookies to access a website and do Task Y) lets say the user visits one of the brands different webpage or even the same page for cross checking.
Cookies are for providing someone certain kinds of info
Can Opal enable or disable cookies when doing Task Y on a non-google website while asked to search that website in the generate block?
Hello does anyone know to how to add first-frame and last-frame for veo 3.1 in Opal ?
Yoo fam
I do have an Privacy Related Question.
minor UI issue, it is expected to be a dropdown since I cannot scroll the content (I need to zoom out my browser to see the full list
"No text generated. Opal has limited quota and it was exceeded. Please try again later."
literally when i started the flow the first time after days and days, i was trying gemini 3.0 pro in my flow, the error is instant
"No text generated. Opal has limited
Please fix the bug. It's only possible to use Generate Gemini 3 once a day.
ig its not a bug
Hi guys. Is anyone else having issues connecting to https://opal.google?
Keep getting this error:/
Can't find anything online if Opal changed into something else?
I just tried a quick sign in and worked perfectly
How strange...It's been like this for me all day
I just resorted to using Ai.studio.google
Did you try in incognito mode? Maybe some extension is causing this
Just tried it. Same result. Have tried today from multiple computers from multiple internet connections and multiple browsers. Even my iphone won't load it.
Ohh weird
Could it be a US thing?
Ohh wait are you in the supported regions?
Ofc it is lol. Then maybe it's some other thing causing it
I’m gonna have to just use Gemini 3 which is working and hope my professor takes it lol
Seems to do the same
Ohh best of luck! 🤞
Ty for the answers
@solid spruce there should be no access issues as far as we know and Opal is still at opal.google. The models are quite overloaded today (google had a lot of launches), but the product should be available! Let me know if the error persists for you.
Heyy it works now! I've been trying since the morning! It must be what you said that the models were overloaded
Thank you!
It looks like Nanobananana Pro has been added to OPAL, but an error is occurring. When will it be available for normal use?
"An unexpected error occurred while calling the model. Please try again."
It looks like Nanobananana Pro has been
Opal not avaliable in my country. Anyone knows when Opal will be released to EEU region ?
Opal not avaliable in my country. Anyone
This flow takes an image and creates a
I also have a flow for Sora 2.
In the Sora 2 option to create videos using your own image, I built a flow that generates prompts as well.
Do you want a course on image generation using Whisk?
This flow analyzes a reference video to deeply understand what is happening in the scene, including action, motion, pacing, camera behavior, atmosphere, objects, and physical realism, and generates ONE single, cinematic, production-ready prompt.
The generated prompt is designed to be reused later with the user’s own face or photo, ensuring compatibility with real images while avoiding any identity transfer from the original video.
So glad I pay ultra and get none of this in the uk
Opal is a fantastic stock app.I put in sources, stock trading and investment-type web sites. Include links to live trading places ,. Enter prompts like this "what pre-market stocks look poised to be a hot trade this morning" ....I ask for noon hot trade updates.. it will blow your mind !.. FYI: like all these mini app generators, not to many parameters, sources (10 is good for stock sites).. just bogs it down . Go for "one-shot" tasks and then build. If you go to far just revert to an older version. Truly one of my fav apps..
mine too
flow test performed on Manus.ia
Flow test performed on CHATGPT
Which one turned out better?
GPT gave much brighter colors, but I also like the mood presented by MANUS
Thank you for the feedback, but the chatGPT only created a stream; it didn't generate the image.
Challenge: I'm going to send another image here, and I want you to guess which one it was made in.
<@&1366831844502274180> This seems very suspicious and they are spamming.
I am preemptly thanking you for addressing this concerns, Mods - Revamp
I managed to create a workflow where it analyzes video and provides a prompt from the video for me to use as a photo of myself in any other AI.
Does opal can read text from (bulk) pdf uploaded in gdrive? and also what is the maximum files opal can handle at the moment? (since there is currently no loop to individually handle files one by one)
these are also so cool
I think it can if it doesn't exceed the context window
Is it a bug where I get signed out of Opal every week?
I don't face this error with any other google labs experiment
i made two amazing apps in opal but now i cant use them due to limits, what do i do?
I do not recommend you do this because it is not 100% functional yet. It is not a complete platform. There is still a lot to improve, especially when it comes to reading PDF documents in general. So I do not recommend this type of workflow, because you will deal with a lot of headaches and still get a poor result, you know? I recommend you test other platforms, because right now using OpAl does not work well. You will face many errors and need to keep adapting. In the end, you will not get the result you expect.
The most I recommend is a PDF. For example, using it as a summarizer to summarize a PDF, you know? Even then, the summary needs to be short and concise. If you make a long summary, you will face the same errors and keep adapting over and over. In the end, you will get an average result or a poor result.
For example, I have a workflow I built to summarize PDFs. It is not a large workflow for this exact reason. I know the platform limits. It gives me a short summary in a way I get a good result. The PDF summary is acceptable, but the workflow is small.
I did not understand. Did I do something wrong?
Yea true indeed, It still works if I used like 15-20 pdfs to summarize and filter CVs at the moment but not more than that. But I think I can have trust on opal about their development to handle bulk processing or large files in the future, we just simply needs to wait
For safety, I recommend at most three PDF files before it starts to hallucinate. If you upload many files, it is not ready yet. In my view, it does not work well with multiple files. Based on my tests, up to three files give a good result. More than three and it starts to hallucinate and mix information. From my tests and research, the highest number with a reliable result is three.
Three files for the best output result, I see, noted, thanks for the information
This is based on my experience and on frequent tests I run, like the example I sent above. When you send more than three PDF files, it starts to hallucinate. The result becomes less accurate or simply not good. In the tests I developed, where I tested and compared each result, the number that worked was up to three PDFs. With three PDFs, it does not hallucinate. This is only an observation from my own tests. I am not speaking as official platform support. This is personal experience only.
If you want help or want a workflow like this, explain roughly how you want it or what you need, and I can try to build it for you.
Here is a spoiler of the next workflow I am going to create.
😲
Give me image ideas to test my workflow.
New workflow I created in Sora 2. You enter the idea for the video you want. You also enter your username so it creates the video with your traits when you use the Sora 2 app.
With the Google app, I created a QR code generator tool.
@versed lotus Question for the moderators. Where is the Discord for this tool?
i made two amazing apps in opal but now
Could you provide me with a link so I can see the limitations?
Is adding events to google calender and wake up call not integrated in opal yet? I thought google calendar integration can be done through API but I'm new to opal also any suggestions about wake up calling feature? (like "okay google" in androids)
Hello, if you are looking for the Google Gemini discord you can find it here: https://discord.com/invite/gemini
hello all, i have question on the dash board... It seems like my ID cant save the work that i have done in the past for the opal
I dont know why after i finish it it say the dash board cant be loaded...
Is adding events to google calender and
hello all, i have question on the dash
I don't know if someone already reporting this issue or maybe only me facing this issue,
Here is the flow
(Signed Out state) -> Open shared opal project link from communities -> Start -> Sign In -> Error Produced
It works fine when we refresh the browser after that, I think something wrong with catching the sign in credentials. Here is some pictures to explains.
Every couple days my app disappears and the only way I can recover it is opening the web history and reloading it, it does not appear in my apps dashboard.
Yeah, I’ve seen that happen before. Sometimes the app just doesn’t sync properly in the dashboard. Reloading from web history usually brings it back. I can show you a more permanent fix if you want feel free to add me @velvet goblet
Great. Many thanks.
I don’t want this four modules to run simultaneously but I need them to work separately. Is there any way I can make them not run in parallel? I don’t want to overload the model
Yes ,that’s possible. You can configure them so they don’t run simultaneously but can still run in parallel when needed by isolating each module and controlling their triggers. I can explain the setup if you didn’t like @velvet goblet
Solved. Thanks.
any way to fix this problem i am in Norway
Opal Bug Report: Unable to create new Opal app in Chromium
When click on 'Create new' button, it just spins forever.
The Console has following error:
/_app/?mode=canvas&shellOrigin=https%3A%2F%2Fopal.google:1 An iframe which has both allow-scripts and allow-same-origin for its sandbox attribute can escape its sandboxing.
action-tracker-B0nJzTln.js:938 [shell guest] Sending establish MessageChannel request to https://opal.google
shell.js:22 [shell host] Received establish MessageChannel request from https://opal.google
shell.js:33 [shell host] Sending establish MessageChannel response
shell.js:35 [shell host] Exposing host API
shell.js:35 [shell host] Connecting to guest API
action-tracker-B0nJzTln.js:950 [shell guest] Received establish MessageChannel response
action-tracker-B0nJzTln.js:927 [shell guest] Connecting to host API
action-tracker-B0nJzTln.js:929 [shell guest] Exposing guest API
main-base-BKQXuGxm.js:12570 [Opal Visual Editor Initialized]
index.js:111 [Opal Visual Editor: Version 1.30.0; Commit 2025-12-15, 10:19 a.m.]
action-tracker-B0nJzTln.js:709 [user graphs] error while listing user opals AggregateError: error while listing user opals
at Ar.mt (action-tracker-B0nJzTln.js:680:35)
at async Ar.gt (action-tracker-B0nJzTln.js:648:13)
wt @ action-tracker-B0nJzTln.js:709
I was able to resolve this issue by giving Opal all the permissions to access my content 😢
Don't worry, we only access your content when you ask us to upload documents or use it for a reference! We do not use your data for training 🙂
this shouldn't happen 🙁 Are you using the same device and browser?
📳 Mobile fix FYI - thanks to opalites who have been reporting mobile errors with sign-in! We think we've fixed this issue and it should also unblock usage on Safari on desktop as well. Let us know if you're still hitting any issues though 🙂
Yess I'm on chrome windows
opal users are called opalites? 🥹
Yes, my fellow Opalite 🫡
Bug Report:
Yess I'm on chrome windows
I build a prompt refine app with Opal, it got a copy prompt button in the end to copy the refine prompt text to clipboard but failed with this message 'This content is blocked. Contact the site owner to fix the issue.'. Who or how to contact?
Thanks @elfin tulip I believe if you could run tests by enabling only one permission you can reproduce my bug report. I am happy to provide any additional information.
Ah I misunderstood your original issue. Opal requires all the drive permissions to run right now. So this is working as intended. We don't actually need all those scopes, so we're working on an update but for now you have to accept all drive permissions to use the prodcut. We only ever access a drive file if you upload it into Opal though! We also store all your Opals in your drive so your data is yours.
I build a prompt refine app with Opal,
This error is common. The platform still does not support button customization or button actions. You cannot add buttons with actions. In my research and analysis, even after many attempts, I could not do it. You cannot assign an action to a button inside the platform.
Hello, Opal is available even in Venezuela, so why isn't it available in Türkiye? #opal-general・✨
We have just released a fix that we believe resolves the error you've been experiencing.
I can log in to Opal on my iPad Pro 13, but I can't “cmd + x/cmd + v/cmd + c” to the prompt.
We have just released a fix that we believe resolves the error you've been experiencing.
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Thanks so much for your patience as we looked into this. I'm happy to report that we've just released a fix that we believe will resolve the error you've been experiencing.
Could you please give it another try now and let us know if it's working for you?
Holy moly, tried notebook LM today and I'm shocked! It's so so good. It's unreal. Thank you for whoever made that
Is an embedding feature in Google Opal's roadmap?
gemini 3 pro not still not working. someone from google can help?
I was able to connect and run my opal things from my phone. It took three attempts logging in after selecting my Google account and providing permission, but I got in. Thanks!
quota is very low for that model, we've requested more but don't have an ETA on that yet. Use Gemini 2.5 flash for more dependability!
Is an embedding feature in Google Opal's
Hooray 🙌 Opal in the Gemini web app!
They keep rolling out cool new stuff 😭 I get distracted very easily and lose track.
thanks! it was a really fun sprint to get it done before the holidays
Really has been amazing to watch the rapid pace of advancement across the board. Congratulations!
Good to see that Gemini 3.0 Pro Image is working now 🔥 . But I do have one feedback, the image resolution is very low as compared to 2.5 Flash image. The downloaded size is around 80-120KB, whereas 2.5 is around 1-1.2MB. I tried prompting it for higher resolution image but it doesnt seems to work. Wonder if there is any workaround? Thanks in advanced,
Opal Turkey?
Btw, Congrats on getting Opal on Gemini!
Okay, I have a few ideas about making an Opal which can help me on a bit of guesswork with Excel/Google Sheet Formulas (like which formula to use. Since, There is a ton of formulas). I dont want to over engineer my prompts. So, Lets say i want to make filter which checks for certain action words amd triggers only 1 generate block. How can i do that?
Update. Gemini is now connected to Opal. Before, I could already build complex flows. Now they will become even more complex.

💖
Hello there, in from Europe, Spain Google, and i try to enter to opal it i dont have anticipated entry, we need to do something as enterprise? W
How Complex are talking?
at this level
I am currently trying to make an Colab Project where i can use it for HTR
Which when support for Colab in Opal is added. I will pretty much use it immediately
Tell me more about this project. Maybe I can try to recreate it in Opal.
I am trying to make it recognize my Handwriting
Which probably wont work. Since the large amount of training data i have to provide and where opal can access it or not. Also, The pre-processing using Nano Banana (No Pro) isnt that great for this project
Oh, I understand now.
I am currently focusing on Creating an Logic Applier Opal. Which can take any logic for an Excel/Sheets/SQL Function i would need and will find the correct function for it
Since, Excel and Sheets have tons of functions (This is an understatment)
quota is very low for that model, we've
Btw, Zauren. Is there anything you have learnt while using Gemini 2.5 Flash?
As the model for your Opal?
Yes, yes. I learned quite a lot. It is a model I personally consider average. It sits around the middle, in my view.
Basically, I only want to use the Base Model in Opal (This is more of a Personal Challenge). So, Thats where i stand. Although Lymia and AudioLM and other models are allowed. But not any of the Pro Models like Gemini 3 Pro or Nanobanana Pro etc
Hi, is there a new info about aviability of Opal, in region like Slovakia? Thank you
New project I am working on to create an AI digital influencer.
Hey! That's awesome 😊
In Opal, is it possible to use APIs?
In the tests I have done so far, I have never used the API, and I do not know if it is possible to use it, because I have never used it.
Ok, Thank you
@bleak anchor
Can you make a suggestion/recommendation?
I think an if statement would be super helpful. Like for determining if the user wants a text output or google sheet output, etc
But overall Opal feels like Google's version of Apps Script but more like Scratch than just code. It isn't quite packed with features like I'd expect from a fully developed Google product, but I can't wait for more features to be implemented
And of course, the AI is extraordinarily helpful, and UI is easy to use
we're working on more complex logic, so this is a good feature request! CC @quaint eagle for viz
alr thank you
but definitely look at scratch.mit.edu for inspiration on things that could be added; they feel very similar, but opal obviously has AI and google workspace integrated, making it more realistic to use in the real w orld
when i chose to save output into a google docs is just overwrite the old one and not creating a new docs into google drive 🤦🏻♀️🤦🏻♀️😭 pls fix the output situation, its impossibile like this
It is working now, but formatting is way / difficult to use in mobile still and the app did successfully generate
When will 3 flash come to Opal?
i'm sorry, i'm newbie. Is there a credit/token limit for generating videos in Opal? If so, how much is the limit and how do I get it?
I'm trying to create a mini app to generate videos using image input as a reference in Opal, but I'm having trouble with the video generator's token/credit. The program works fine, as I expected. Please provide information.
When will 3 flash come to Opal?
How does the Code Execution API work? Where would i give the code to execute?
What languages can it do code in?
interactive controls for gathering data from AI generated HTML are very useful (like select/checkbox/range ), so AI can decide the next move
here is the doc https://ai.google.dev/gemini-api/docs/code-execution I also create a demo mini-app https://opal.google/?flow=drive:/1mnxaJXxHuEHpnFZiZbtpNy2-PiVg6wu9&shared&mode=app
Thanks. Your Demo Mini-App is cool
Hi Opal says "Opal is not available in your country yet" but Finland is supposed to be one of the supported countries
Hey, Opal is not yet available in Finland. The list of countries in which Opal can be accessed is here: https://developers.google.com/opal/faq
Do you know when it will be available ?
this
Hi all. Just updated my opal workflow with Gemini 3 Pro and now it throw the following error: "No text generated. Opal has limited quota and it was exceeded. Please try again later." I haven't generated anything in awhile so I can't imagine I'm over the credit limit, which does anyone know what they are or how to get more?
Does anyone know if opal quota is higher for ultra users
Hey so I'd love to make a slight fork to the Gemini "Storybook" Gem where it always uses the same image reference for its source characters...
I can't find a way to copy or edit or have pre-promted it, so alas I'm here at Opal. Would Opal be the right way around that?
I effectively want the same app as Storybooks, just with image of character and names of them pre-linked.
Hey so I'd love to make a slight fork to
You can expect this feature to launch in a matter of weeks, potentially even sooner.
Does anyone know if opal quota is higher
I'm trying OPAL for the first time to create a Mini app to summarize some text from a website I own, but i'm having sort of DNS errors, this website is live and running. DNS is ok. Does anyone know what can be the problem here ?
Ok I get it. It's not possible to request using python but with the component Get Webpage it works very well. These mini apps are insane! Thanks for who created this feature!
I'm trying opal, i'm writing something that is supposed to check for character consistency in a fanfic i have, the model runs perfectly until it's time to produce the HTML, it brings up random information from the internet instead of using what's in the console, also opal uses flash 2.5, is there a way to change that?
Is there an ETA on when we'll get Gemini 3 Pro / NBP availability. I understand not wanting to give it away, but, at least the ability to pay for it via the API, or, use our paid account?
I am trying to get a custom gem to connect to my local system but can't find the advanced editor page to add the "custom tool registry" or "import open API"
hello! I was just trying out Opal and noticed Gemini 3 is available in most options except here under advanced settings
I'm completely unable to use Gemini 3 pro even for one sub task in opal
I'm told opal has limits for the very first time I try it
On both a pro and ultra account both personal
Hi! Gemini 3 Pro is very limited right now, please use the 2.5 models! We added it as a preview model, but it's been very overloaded. We're aiming to get 3 Flash quota in a few weeks, so that will be the best way to access Gemini 3 in Opal soon!
hey are you able at this time to give any word regarding some users accounts becoming ineligible to use with antigravity seemingly at random?
i was able to use it fine since release on my ai ultra account (personal not business)
now i am being told my account is ineligible
lot of people on x and reddit with the same thing
some believe it can be caused by logging in to Gemini Code Assist extension. I can say that i did indeed do this a couple days prior.
Not familiar with Antigravity user access, they should have a separate community forum though?
I'm trying opal, i'm writing something
there is none as far as i am aware
only a reddit
@elfin tulip
I am currently building an automation workflow using Opal and integrating it with a Telegram bot.
I would like to know if Opal supports automatic execution (Auto-run) when a flow is accessed via a URL with specific parameters.
Currently, I am passing data to the input node using a URL like this: https://opal.google/?flow=...&final_keyword=MY_KEYWORD
The data is correctly populated in the input field, but the user still needs to manually click the [Start] button to begin the flow. Is there a specific URL parameter (e.g., &run=true or &autostart=true) or a setting within the editor that allows the flow to execute immediately upon page load?
If this feature is not currently supported, I would appreciate any advice on alternative methods (such as Webhooks) to achieve a fully automated experience without manual clicks.
just code it
Got around to trying Opal in Gemini. It's great!
I am currently building an automation
😂
guys in my country, opal is not available sadly. Is there any way i cant enable it?
Hello everyone!!!! 🥳 🥂
anyone had succes generating python code wit this?
i get data fetch request failed when i try to run anything other than"generate python code : hello world"
Hey, 💖 Opal team! Any best case scenarios for Opal as a super gem in the Gemini app?
Hello can anyone help how connect output result with email ?
Hello from Spain we are waiting to start, do you know when we could use it?
guys in my country, opal is not
I am trying to perform image-to-image conversion in Opal. The output is fixed at 1376 x 768 pixels. Is there any way to change the output resolution to 1920 x 1080 pixels?
Hello can anyone help how connect output
Is there any way you can publish opal but without giving an option to remix or see the prompts used ? Or api which will use trigger opal flow through another code base ?
Is there any way you can publish opal
I can’t open opal just loading
Please tell me
Sometimes 400 error
Or not in India
It is available in India. So, Idk why this should happen
hey what are the exact daily quota for image generation on opal
I am trying to add a NotebookLM as an asset to an Opal Step. Gemini has given me the suggestion of - Inside the Step Prompt text box (where you type the instructions), look at the bottom left or right corner of that specific text box for a small "+" (Plus) or Paperclip icon. This icon is not visible. Any ideas on how to add a NotebookLM to an Opal Step?
It isnt possible in Opal as of now, I believe. I dont see that option
Thank you for assistance but not fixed@dire kiln
Thanks for the response @Revamp. I wonder if Google has this NotebookLM asset in a "feature add" timeline? Would be so cool and make agents SO MUCH smarter.
Gemini 3 Pro is very limited right now, Since then, why not quickly support gemini 3 flash, thanks
Gemini 3 Flash is currently overloaded, As i heard. (I am not staff)
It would be really cool. (I am not staff) So, Idk much about whats on the timeline
Can we use AudioLM to IPA Pronounciation into audio and 1 Audio for Each IPA Pronounciation? Like, I want to be able split the audio files for each word and not be stuck with 1 large audio file
Or Can we say the Output Step to only play from a certain timeframe of the audio when pressed an button?
Merry Christmas to Everyone here
"I'm building a Furniture Part Calculator for built-in interior work. Has anyone tried using Opal to create a mini-app that generates cutting lists from user dimensions? I'd love to see if I can automate the workflow further."
am trying to build a app for my news posting project but i dont have enough coding knowledge anyone can support me
i attached my first try
Google Staff pls add estonia to opal i cant wait anymore..
Hi
Hi, anyone know when Opal will be available in Europe?
Opal currently only generates images with a 1:1 aspect ratio. Please, I really need support for other ratios like 16:9 or 21:9!😭
Does Opal have a feature similar to ComfyUI's 'Mute' or 'Bypass' nodes? I only want to test specific parts of the workflow. Running the entire flow every time is slow and wastes my credits.
maybe it would take a lot of pressure off the servers, too.
I can’t open opal just loading
I am trying to add a NotebookLM as an
Can we layer multiple iframes on top of each other like an layered map which are all just Google Chart Geo Maps and control which layers i want to be visible via buttons in Static HTML on Opal?
Is opal going to support webhook client and server ?
It is an highly requested feature. It is on the roadmap according to the staff
It would be very awesome to see opal implement customizations and features of n8n.
I am very interesting see how Opal is going to address Multiple Outputs (A Looper Function which is also confirmed to be on the timeline), Mobile Capability, Customization, Integration with Other Tools
Hi, anyone know when Opal will be
Opal is capable of generating in different aspect ratios. See here:
Multiple outputs are achievable via parallel node structure
Muting nodes would be a HUGE plus
How can we check for Deep Research and Canvas Usage Limits for Opal?
Yo
just waiting for code wiki
Can we use opal as we are building in blockchain
I have spent weeks fighting with Google Antigravity, it keeps generating broken websites and can’t handle even basic integrations with Stripe, PayPal, or Portals for my business, and now Google won’t even reply to my emails or messages. can someone help me? Thanks
Feature Request For Gemini chat parity, enhance the "Add Assets" feature to include NotebookLM & Attach Code(and/or github repo) to help centralize assets.
Feature Request For Gemini chat
Mallaiah, What are Quota for Deep Reserach and Canvas Quotas for Opal as of now?
!!!!thank you ill try!!!😭 😭 😭
Hi i wanna ask, will the Gemini 3 Flash be available anytime soon on Opal?
Hi i wanna ask, will the Gemini 3 Flash
Anyone know when Opal will be released in Sweden?
No ETAs for Any EU Countries Yet. They are working on it. (I am not staff)
Ok thanks for info👍🏻
hi. this is a very basic question. are there any comprehensive pdfs (not videos, not interactive tutorials) that get into the nitty gritty of opal? i'm having a hard time finding old school materials 🙂
maybe it's not that basic a question after all 🙂
Yep, There isnt as of now. It is relatively new and still in it's experimental phase. Maybe 2026 might be the year where users might create an comprehensive PDFs
ah ok. trying to bring chatgpt up to speed on it to become an expert 🙂
and it likes to digest text/pdfs
Use NotebookLM for some of your research, I find it's been helpful for this exact use case.
Anyway couldn't share this in the regular gemini image - gen channel but figured someone might get some utility.
gemini 2.5 Flash Image gif creator using a simple "2 seconds later" stye prompt @ each img gen node to create a progression. Outputs converted to gif via python execution and all output to final markdown page (for speed)
https://opal.google/?flow=drive:/1nnqzXE2yU5liaDcWYjX_GuN-gyMbklC7&shared&mode=app
Feel free to change the parameters in the editor (3 pro image vs 2.5 flash image, gif speed / filename, etc etc)
This workflow highlights a simple concept of creating a batch of images from scratch and processing them w/ a "code node" (my label not Opal's)
Feature Request for Gemini API - Add ability to call Opal for AI-assisted output. Allows ability to factor out context in Application/Google AI Studio into opals. Read is most important then Insert/edit/delete could come second but likely the existing UI for that.
Feature Request for Gemini API - Add
Thread
Please bring it to Greece. I am paying for Google pro too and I don't have access 🥲
Please bring it to Greece. I am paying
Hey, does any one meet Opal's limitation when using gemini 3pro?
Woops, Opal is not allowed in Korea. Is it real?
PLEASE add the ability to set the default state of a GEM to the current gem. Even if you just accidentally type something into the Edit These Steps, it overwrites all your work with some prior version
Please Use English and not other Languages. It is stated in the Guidelines. Discussions are only supported in English as of now and not Other Languages (Not Staff)
hi all
I was trying to build email integration but that is not supported in current opal
does google have a plan to add marketplace to opal? I'm interested in building extended support in opal
Really Interesting
I was trying to build email integration
opal cant generate side&aspect=21:9 image,but gemini can
its relly cool to generate gif
its output that I ask Ai to make recursive reasoning. can I know opal supports any loop features?
Currently, A Looper Function is under development and might release soon. (News by the Staff). It can be able to do what you need to here
its output that I ask Ai to make
Is Opal still only available in the US?
No okay found a country list now
I told the premade recipe creator app in the Gemini+Opal integration that I had carrots, whipped cream, a half eaten can of Spam, the 😋 emoji and a Samsung 65" 8K Neo QLED 2025 AI TV
Unfortunately it only made recipes with the carrot, whipped cream and spam. No emoji or TV 😢😢
i dont have quota to even finish 1 blog example with 2.5 flash... is this exepected...
how to get more quota
How big is your opal setup?
Is there a way to use opal without letting it download all of your google drive?
Cannot sign in opal.. why?
Cannot sign in! please help me out!!
oh.. problem sovled.. all of sudden
is there any way to extract opal project?
like after the "share url" can we deploy it?
Why don’t my opal work to generate video from an image?
Share you're link.. let me take a look
Why don’t my opal work to generate video
Is there a way to use opal without
@grizzled girder is there any way to deploy or download our project? I'm kinda new to opal and wanted to know if final result is the link itself
Is there anyway to get Gemini 3 access in Opal? I upgraded to AI Ultra hoping to be able to use Gemini 3 in one of my flows and im still getting the "No text generated. Opal has limited quota and it was exceeded. Please try again later."
Thats what happens when you trust AI LMAO i asked gemini if AI Ultra gave me access / increased tokens in opal it told me yes 🙁 pesky pesky gemini
I'm also getting the "No text generated. Opal has limited quota and it was exceeded. Please try again later." issue across separate accounts.
I think it's a service issue not an issue with our workflows or plans.
I just read up a little and figured out that Gemini 3 doesnt really work as the quota is still so low. and AI Ultra does not increase credits yet but soon will allow for more videogen. Shoulda hopped in the discord first lmao
Afternoon! I just found Opal and have spent today playing with it and building a mini-app. I am working on an interview coach. The idea is that you put in the job description, your resume, a URL (public) for the company, and a description of the type of persona you want the interviewer to be (search firm, Initial HR screening, manager of a team, VP of a division, etc). Then, it generates questions for you to answer. After the interview, it reviews each of your Q&A pairs to give you a grade and suggestions on how your response could be improved. I can get to the interview process and that seems to work great. The problem is when I hit the Review Interview step. I tell it to use the Conduct Interview step info (which is the previous step), but it seems to be just making things up that are not even close to what I was asked or how I answered. I know that was a lot. Suggestions? Thoughts? Questions to help clarify?
Hi to everyone. Im hoping someone can help me activate Opal. I have two Google accounts/profiles. A personal one and a Google work one. I cannot activate Opal either on the offical web page Google.opal or inside the Gemini gems. It is very frustrating. I live in the UK. I know rollout has been slower to the EU and UK but I understand Opal should now be available in my region. Can anyone help? Many thanks. Chris
Literally blew my whole day on Opal 
Any one else find the generate web page output to be very frustrating? I am passing it a article to turn into a webpage, and it sumamrizes / rewrites the article before posting it. I've rearranged the prompt like 5-10 times it loves to sumamrize...
Im search samz9182
huh
You know ? samz9182
Hello sam
Sam you there?
Try using a database to store all that interview information, Supabase is good, and have it retrieve that information rather than trying to remember
I would like to be able to print the web output generated by Opal app. Has anyone be able connect and print?
me too
im getting this error
im getting this error
Any one else find the generate web page
I would like to be able to print the web
I wonder if I can upload a mp4 video file as input for making a sepecific app?
I wonder if I can upload a mp4 video
Hello Opal Team,
I am a Game Designer and UI Designer , and a dedicated power user of Opal.
I am currently utilizing Opal Agents to build a comprehensive pipeline for a mobile point-and-click puzzle game (inspired by titles like Rusty Lake and The Room). My workflow covers the entire creative process: from world-building, outlining, scripting, and dialogue to scene composition, puzzle design, prop creation, and cinematic direction.
The specific outputs I am generating include scene art, character designs, prop assets, Unity Prefab construction files, interaction JSONs, storyboards, cinematic stills/videos, and voiceovers.
I have already made significant progress and achieved some great results.
To be clear, my goal isn't to rely entirely on AI. Instead, I want to use these tools to free up my time so I can focus more on the player experience—specifically on crafting hand-tuned Unity animations, polishing every line of narration, and refining mini-game mechanics.And opal helps me a lot.
I also want to mention that this is hands down the best product I have ever used. I truly hope to see Gemini continue to get better and better!
Thank you for google labs
Here are a few suggestions and bug reports based on my experience:
1.Console Bug: Sometimes, after I modify the preceding prompts, the output Console gets stuck when collapsed and cannot be reopened.
2.Node Wiring: The "Add assets" node behaves differently from others; it doesn't allow me to select the connection point positions. It makes the graph wiring a bit messy and less organized compared to other nodes.
3.Node Muting (Critical): I strongly suggest adding a Disable/Enable function similar to ComfyUI (this is very important). Being able to right-click a node to toggle it off would be a game-changer for partial testing, saving us from having to run the entire flow every single time.
4.UI Customization: It would be fantastic if we could create custom UIs to fill in input sections via Enum dropdown menus, with the option to toggle specific inputs on or off.
- I really hope to see a 21:9 resolution option for Gemini 3 image generation.
the UI maybe like that
Is this for those viral glass fruit ASMR videos?
hello guys, is opal can translate data to xls file?>
You can ask Opal to create an app that exports a sheet to your Google Sheet, and then export the file as an .xls format from Google Sheet.
also can i ask opal to make link and record automation on it?
thank you light
Hello Opal Team,
I think it would be good to add Google Pro and Ultra to increase Opal’s limits, since the current limits for Gemini 3 Pro and NanoBanana Pro are quite low.
I think it would be good to add Google
You can create interactive buttons which properly work? (I tried an copy button which didnt work)
yeah,it works
🎆 Happy New Year everyone!
Happy Belated New Year Wishes, Dimitri! I am excited to see what y'all are working or planning to do
Can you help me with making an copy to clipboard button (i am using mobile)?
Btw, If Staff are seeing this? When is Mobile Web Version going to get Editor Mode and Opal Properly?
Hi, i'm trying opal for the first time, and when i try to use Gemini 3 pro in any node i get limit error even though this is the first time i tried using it, and i have PRO subscription.
Also i don't see Gemini 3 Flash available, only 3 Pro is there.
does google have a plan to add
Gemini 3 Models have been having insanely low limits since its launch. I heard they are going to launch a plan which will fix this issue
@hazy tangle this is my source
Afternoon! I just found Opal and have
Hi to everyone. Im hoping someone can
Btw, If Staff are seeing this? When is
Hi, i'm trying opal for the first time,
hi. this is a very basic question. are
2.5 flash is totally enough in most of the time
and 3 Pro is significantly slower than 2.5 Flash
its too hard to use at my mobile 🙁
I want to rename the output files to match the node names in the console. Right now, they are just named 'file0', 'file1', etc., which makes the sorting very messy, and some of the files are even empty.
Ik that. I am using chrome desktop mode
on Samsung Galaxy Tab, Chrome Beta, dragging a box and some fast fingers swapping, the box moved to infinity. I couldn't reach it anymore. Display gets colors glitchers... Containing icon, zoomsk nothing gets the box, now neither any of the boxes nodes. Its just endless void... Just coz I got my Opal done and HAD TO 'pixel perfect align cards' hahahaahha https://opal.google/?flow=drive:/1kzj4hpvfN_3u4jq-L0SAULJc3hqqOpcF&shared&mode=app
On the PC side, the node box may also move to a faraway place unexpectedly, but a 100% reproducible method has not been found yet.
Idk what you are talking about? There is no wallet here and this is Google Labs Server btw
Did you get an copy and paste function working for one of your projects? If yes, Can you share the bits which ya did use for it?
"Hi experts, I want to use Opal to create a workflow that can accept both a Ph.D. dissertation and the detailed thesis requirements. The workflow needs to split the dissertation (since they are usually quite long), and then check for formatting errors based on the requirements. Ideally, it should also check for typos and awkward phrasing. It needs to highlight the issues in their original locations within the text, explaining the specific error and how to modify it. Finally, the results should be consolidated back into the original document format, just with the added error markers and suggestions. Can this be achieved using Opal? For tasks like splitting the text, do I need to use other programming languages or call specific packages, or can it be implemented entirely through Opal?"
Opal doesnt work like regular coding apps. It depends on how you want to separate the dissertation and sample data as an reference for the AI to detect such formatting errors and you can tell it to highlight the errors
I want to rename the output files to
Hi! How often do Opal's quotas refresh?
Since yesterday morning, I have been getting "No text generated. Opal has limited quota and it was exceeded. Please try again later."
I am new to Opal (but a very heavy AI user) and was surprised to have run out of credits very quickly, especially given my Gemini Pro subscription.
I heard they are working on a plan to fix this issue and Gemini Pro Subscribers might already have this added to that plan. (I am not staff)
Thanks! I was surprised because nothing I was doing was especially token-heavy.
Btw, Use Gemini 2.5
Gemini 3 Models Quota are insanely low due to high demand at the moment
Oh, that must be it - I did switch to Gemini 3. Thanks so much!
Hello, experts here! I am pretty new with Opal and am seeking help for my first Opal project. I want to add a weekly trigger to autorun my app and export the results in a Google doc inside a specific folder on my Drive. Gemini told me that both are possible (using a scheduler for the first task, for example). I tried many things Gemini told me but none of them worked. So I want to ask if these features are available but I haven't figured out where to find them, or Gemini has been lying to me. I will appreciate any help very much!
I can certainly say, Gemini have been lying to absolutely. There isnt a scheduler step at all
You can although export results in a Google Doc. But, Control over where it stores? No Idea
No worries!
Thanks much! Is there any other way to engineer autorun in an app?
That's the problem I am facing. The Google Doc is saved in my root folder, not the subfolder I selected.
not available in italy, how else i can generate my app in a similar way ?
Yes sir
Hello guys
Supp bro
supp bro
Yoo bro
Supp bro 🔥
man, you everywhere
Lets goo to level max bro
what lv you right now?
Seems like always interacting here
ahh i see
Glabs bro haha

Hello
Hmm? Currently, I dont know. Apps Script is definitely the best option for this as far as i am aware.
You should probably make an feature request here for to ask the user where they want to store the outputs of AI and What do they want to name them?
Greetings, Staff. Can you explain how does the Plan & Execute Model work and what do i need to provide for it properly enter in a certain sheet based off month (6 Different Sheets, Each Containing 5-6 Columns to enter) and only enter the needed information in the next row after an non-blank row in that sheet with storing a certain information in each cell within a row?
I have never seen it anywhere else. So only, I am a little confused
Thanks for the suggestion. Who should I request the features from? The manager of this channel?
Just do it here. They will create an thread and will respond to you
Dear staff, I would like to request three features that seem unavailable in the current version of Opal:
- The ability to autorun an app, for example by adding a scheduler function to a node.
- The ability to use variables in the output file name, such as today's date or any variable passed from an intermediate node.
- The ability to specify which subfolder of Google Drive the output files (docs, sheets, slides, etc) will be saved to. Thanks!
GOOGLE im begging tou, PLEASE ADD OPAL TO ESTONIA
<@&1366831844502274180> Idk if this allowed and i am suspicious of this person? Can you confirm and have a look at this person
Thank You, Mods :] - By Revamp
Alongside these ideas,
Can we add text boxes which we can config to any node or a task or about how should it output.
Sort of like Co-lab for improved readability and selective editing from the AI when asked to modify
when is Opal going to be available in Portuagl 🇵🇹 ?
its too hard to use at my mobile 🙁
GOOGLE im begging tou, PLEASE ADD OPAL
My Opal project crashes with an "Aw Snap" in the browser when I upload about 6 pics for it to do it's magic on
T-shoot with Gemini, tried some Chrome tweaks to fix it from going over 5gb in the browser, but it constantly hits that limit (out of memory)
When is Opal going to be available in Italy?
When will Gemini 3 Flash be avaliable?
Staff, When can we expect our first update on the one of hundreds of feature requests
Any one (you dont have to tell which one)
Staff, When can we expect our first
When will Gemini 3 Flash be avaliable?
When is Opal going to be available in
Where are you from? Like the EU?
From Asia
Same problem I'm located in Serbia right now
For EU Countries, It isnt available yet. They are working on it. No ETAs (Not Staff)
I am from Asia as well. It works well? Idk whether your countries is part of the support, you can ask the staff for support for your country
Opal can be a great stock trading app...Add sources such as top stock trading sites.. Great prompt "xyz , bull bear trading scenario for a hot trade this morning " We have also added premaket and post-market data as source for output.. use the output with a grain of caution . but sure gets your thought and ideas percolating....Its a mini app so keep it "one shot" prompts. Tip: our COT...(chain of Thought) Gemini 3 pro, deep research a stock company. Take that info put in Notebooklm (super fav).. Now I have the data and I am ready for the Opal generation.. Give it a cool name ! party on! Please share your ideas on this.. and cool stock market prompts .
sorry for the late replay.
its a remix of demo, which shouldnt be too big...
ps: the image is taken just minutes ago
If you did use Gemini 3 Models. This happens, Use 2.5 Flash
Please convey in English. Support is provided only in English in here <@&1366831844502274180>
Well that depends if you can speak English?
Hello! As stated in #start-here・📍; currently, our team can support engagement in English, so the discussion in this server will largely remain in English for now.
Its nothing personal, Rules are Rules. You can create a new opal button and you can create steps which you use prompts to tell what to do
Next Time, Please Convey in English @silent sierra
how can I use OPAL for creating long-form video as an audio content to upload Youtube?
I've made norde but it said "limited to create video."
you want a podcast to upload in Youtube or you have an audio to create long form Youtube Content?
Hi there! Id like to have a registration in OPAL in Ukraine. I know, that it is already works in my country, but it gives me the denied answer. All the settings of my google account Ive checked. Can you help me with that?
I can't, I am not staff. They will contact you soon, I believe
Thanks👍
I want to create a podcast!
and I already use google ultra plan in flow site
Hi there! I`d like to have a
You can create a podcast with Opal, here's an example from the gallery: https://opal.google/?flow=drive:/11AW456rFHTPw198bzbeysTRlkIgMas9U&shared&mode=app
but when i get a message "limited to use opal to create text", how can I keep going?
but when i get a message "limited to use
Yes, You can also use NotebookLM if you really wanted to
They are working on EU Countries. No ETAs as of now
i'm using it in Tunisia and it works fine .
Thanks for that idea, Megan Li
@dire kiln yes, but I don't like NotebookLM. I find it both annoying and stupid
Then, Opal is the best way
Feature Request: Ability to use Stax's Evaluators directly in Opal
As far i am aware Tunisia isnt a EU Member? EU Member States will recieve Labs Experiment the last
Any suggestions for a beginner opal project? I tried to make a video generator from the document but it made a 7 second animation of the title of the document alone... So please help me 😭
The worst thing is, even when it is available, you are supposed to be above the age limit... I broke my head trying to figure out why I got that message even though my country was there on the list 🤣
What do you do for a hobby or living?
Find what that is and see what you can do from there
I'm 15... I don't have a job... I just do this for fun.
Interesting, I am more over doing an vocational in Desktop Publishing and doing Communication Design in my spare time. It isnt my main field. But, For AI related stuff. Its really Interesting and helpful
Cool
Like what else you got as hobby other than working with AI?
I'll have to try Opal.
My first atempt to run Opal; a test of the blog post writer in the example gallery returns "No text generated. Opal has limited quota and it was exceeded. Please try again later." error - I cannot run anything, even though I'm logged in with my Google AI Pro subscription. Am I doing something incorrectly? Anyone else encountering this??
Something went wrong. Please try again or send feedback. Why am I constantly getting this error, friends?
Can you may be send a screen shot of the error message and your Opal Setup?
Morning Brief its insanley awesome.
Opal -Gem .. Makes a great morning brief.. Tell it what subjects you want to know , what time , world news, stock news, book news etc.. I ground with my source URL's ..Ie New York Times best seller list. Reuters stock reports, local events, concert announcements....etc
Can you share the setup ?
yeah that sounds like someting i caan use instead of missing out on top headlines
Did you meant to post in #flow-videos・📹 or #whisk-gallery・🖼 ?
Is opal now connected with gemini as super gems ?
Hi @hoary cargo videos aren't generated in the discord server. In order to make a video you'll need to go to https://labs.google/ pick the tool you want to work with then go from there. If you need help with prompts please ask in the appropriate channels for advice. Thank you!
I’m blown away by Opal, it really is no code, certainly not perfect but so much potential, is there a best practices website or something?
Currently No
Ooh something I could start!
Okay, Then
Something went wrong. Please try again
What do I need to do to use it in Türkiye? I came to Türkiye for work and I want to use and experience this AI, but the system recognizes me as living in Türkiye and doesn't accept me. What should I do?
Use a VPN, If thats the case
Is there a way to share an Opal app without exposing the editor?
Nope, Requested Feature as of now
Currently, They are working on making the Google Docs, Slides, & Sheets work properly
@grizzled girder, Can we expect markdown (.md) as an output format soon? I am waiting to see, when NBLM can be added as assets.
Btw, Can we connect multiple assets to one step (I am doing for the output step to keep the output step consistent)?
I was extremely excited to try using Sheets input for this but it looks like it's not actually working yet. Any timeline on when that might be ready? Just gives "internal error" no matter what I try to do with it.
hi! can you share your opal? Sheets input should work!
yes, you can connect multiple assets to one step! and we're working on NBLM as an input/asset integration. Hopefully will have something in next few weeks here for you to try
i stripped it back to only the row lookup but still erroring when i try to run even this simplified version
https://opal.google/?flow=drive:/15mP8-3Mi0_O-RCjhRsZlfk9nOCmLRU1-&shared&mode=app
if it's something with the headers or formatting i probably can't edit the sheet. this is wired into other apps that would need to be changed with it
would be nice to get a little more detail in the error
i stripped it back to only the row
guys i also should point out you have a link here in your reference doc that leads to nothing because it doesn't actually seem to be an invite. so if you're not already on the discord it's confusing
thank you for flagging 🙏 we'll get this squared away!
Any marketers on here working on AI Workflows want to chat?
I’m blown away by Opal, it really is no
i don't really know where is the go to product sometimes when I have an idea for these 3:
- Gem
- Opal
- Build in AI Studio
Gem looks like just to be an agent for mostly the default outputs in Gemini, but that there is the Storybook example that looks like something in Opal (but it has been there before Opal was launched) and I cannot reproduce the same result as Storybook with normal Gem.
Opal looks pretty much the same as Build in AI Studio, except you have a user interface for the logic flow but you cannot download the code like Build in AI Studio, but who needs a user interface if you can just prompt the change😂
Is there an official explaination for when I should use Gem, Opal and Build in AI Studio?
I'm pretty sure that is a common question for many of us. Below is my best summary of each, but feel free to chime in if anyone can better explain it
Gem = simple agent for generalized output for a particular context = personal assistant
Opal = complex agent workflow for specialized output around varying context= Specialized Doer
Build in AI Studio = Specialized Orchestration/Developer Agent waiting to turn your ideas into reality(or web apps) = developer
If you were using them all together, you might start with a Gem to expand your ideas and to create as an expert in a field you want to build an app in. Then, you would work with it until you get to your MVP workflows necessary to reach a specific milestone/goal. You would then create an Opal with those MVP workflows to create that specialized workflow/output for your MVP. Once your Gem is complete, you can use it over and over to generate similar output(think of it like a content generator). Finally, your Google AI Studio, might be used to compile each of your Gem's outputs/context into a comprehensive "Web app".
Perhaps one day, Opals will have their own API and you can dynamically call/use them from within Google AI Studio.
Greetings, That sure helps with generating the output step a lot faster! Thanks
I'm trying to use Gemini 3 Pro on Opla, but I keep getting this error saying the quota has been exceeded.
Any idea when it will be available again? 😭
Would really appreciate a higher usage limit!
Hey all, Wrote a post about Opal, would people take a look and share https://www.punch-tape.com/blog/how-marketers-can-build-ai-tools-without-writing-code-building-apps-with-googles-opal
Also, I'm going to create a virtual event around no code tools for marketers, is there someone on the Google team that could speak?
hi ladies and gents, I just tried to use opal for the first time, it says it's not available in my country yet 🙁
please dm @versed lotus !
having some real issues with the "generate spreadsheet" action in opal. First, it kept updating the same spreadsheet, instead of creating a new one. Now, it just creates an empty spreadsheet that says "Here's the new spreadsheet". Here's my prompt for that step:
After writing the data, display the direct URL to the new Google Sheet tab as a clickable markdown link: Click here for your Asana Import File.
(also includes the context from the previous step)
having some real issues with the "
hi ladies and gents, I just tried to use
hi, when will this come to Emea?
When will it come to Sweden?
Maybe you should find a VPN of another country
Feature Request for Opal: Conditional Step & Named Groups. Where we can enter if, ifelse, for statements and it only runs or executes the prompt for the condition provided. Also, Error Outputs is also necessary for named groups or individual steps
Sort of
Really Handy when making multi-functional Opals
Hello
I read it. Pretty Interesting. I am not an marketer by any means. So, I dont know the limitations and situation of marketers
Is there a way to export the backend prompts and connections in order to publish my app on github? I've already published it in Opal, but I'd like to have a way to share it for my portfolio.
You can put the share Opal link and connect with an Button to send the user to your Opal App
Feature Request for Opal: Conditional
Hi team earlier there was option to connect mcp server with opal in integrations in global settings not visible now any idea was this feature removed or just a bug
Hi team earlier there was option to
Hey there! I'm new to Opal and have a quick question. I'm using a markdown file with some images formatted with when the links are rendered, the image is blank. I'm using direct links. Any ideas on how to get images rendered?
Hey, I Can't login in Opal, anyone help me?
An unexpected signin error occured
Error checking geo access
Are you using a VPN by anychance?
Maybe Refresh our page or your browser. If neither works maybe restart your device
Yes I am using VPN but with residential IP...
AH! I need Opal!
Maybe turn off that VPN? VPNs randomly causes issues with Opal
Which country are you using Opal from?
China
Interesting. Let me check if you guys have availablity (I aint staff)
Frequently asked questions and best practices | Opal | Google for Developers https://share.google/WAXURnWmnXaef0R5F
Unfortunately, Opal isnt supported on China for some reason
Same for Netherlands?
I dont have access to Opal too, Im from czech repub
Yes, The Website shows all supported countries
Link?
Here
Yea
Hi
What would be the reason ?
I get an error on a json upload input when i try to run an extraction step. it's a long file. do we know of a line or character limit on json assets?
Hello! I’m an active user of Opal based in South Korea.
I have two questions regarding usage limits:
- I am currently subscribed to the Gemini Google AI Pro plan. Is Opal’s usage limit independent of the Gemini plan subscription?
- While designing several workflows, I received a message stating that I’ve reached my limit in Opal. If it is indeed separate from the Gemini plan, is there a way to purchase additional credits or subscribe to a dedicated Opal plan to increase my usage? I would like to explore more features and build more workflows without these restrictions.
dude please add india, estonia to opal gng im begging atp Crying_Cat
India has been available to Opal. Since, Diwali
oh but yea thats the problem it says opal is not available in your country yet
The answer to the first question: No, As of now. That might change very soon tho.
The Answer: Currently No. Please Switch to Gemini 2.5 Flash or 2.5 Pro. Gemini 3 Models have Low Quota at the moment. Which the Opal Staff is planning to introduce new plans which will definitely connect with Gemini Subscription
okay
Do you use VPN?
I am from India as well
I dont?
do you recommend me one?
I don't know any good one tho
ohh okk
Are you not from India?
lemme ask gemini
i was born there but now im in estonia
Oh
Estonia wont have access. Since, Opal hasnt figured out the legal side to EU Countries as of now
okk
okkk
Has anyone ever embedded an opal work flow into a platform, website, app, or added a payment gateway?
Also, does anyone know how to get gemini 3 to work on workflows? I tried to switch the llm in nodes to Gemini 3 but it gives me errors
Nope & Nope. The Quota for Gemini 3 are super low due to demand
For the First Question. I am in the planning to implement it into my Personal Website
From IND
@dire kiln thank you for your response i appreciate it. That sounds cool
hi, when will this come to Emea?
When will it come to Sweden?
What would be the reason ?
Hi, I’m currently using a Google Workspace Student account (school-managed account).
I’m getting the error: “You do not have access to Early Access Apps.”
I’d like to ask:
1 .Is this because my account hasn’t been granted access to Opal yet, or
- Opal does not support Google Workspace Student / managed Workspace accounts?
If it’s option (2), could you please explain why Opal doesn’t support Workspace Student accounts (e.g. Google admin restrictions, Early Access Apps policy, or security limitations)?
Thanks a lot 🙏
Did your Admin allow you access? (Not Staff)
how can i access opal ?
This isn’t an official answer but I also use university controlled apps. The reason for this might be because it’s a school administrative controlled account any student access to apps is controlled by the school admin themselves so the enterprise account as a whole might already have access but certain apps are restricted for the students
how can i access opal ?
Hi, I’m currently using a Google
Hello why i can use every other google product in turkey but not opal ?
Has anyone figured out how to get the web layout to add more padding to the bottom for mobile? This is a samsung phone with a chromium browser. No matter what I put in the step it won't pad enough to make the submission field visible.
The main reason is that the Google Opal service has not yet been launched in Turkey (nor in European Union countries) due to issues related to legal and regulatory compliance. While other Google products are available, Opal is still in a trial phase and needs to comply with local laws before being launched there.
what this means "No videos generated. Video generation with minors is not supported. Please try a different prompt or send feedback."
That means Google Nano Banana's Safety Feature was triggered
Hi! I’m one of the people who started using Opal in Korea.
Right now I’m trying to generate 20 images, and I’m currently matching 20 separate Generate (image) nodes one-by-one.
In Opal, is it required that Text → Image is always 1 node : 1 image?
Or is there a way to do batch generation (like one node generating 20 outputs, or a loop/repeat)?
I also want the 20 images to be saved separately with filenames numbered 1–20 (or 01–20).
Is there a built-in way to auto-number outputs, or a recommended workflow for saving each result as its own file?
I’ve been working on this all day and still can’t find the answer. Any guidance would really help 🙏
what? the video has no relation to any safety things
Mb. Then, You can explain your opal to staff and then, they will tell you why is this happening and they will probably fix it soon
who shall i dm
is 8 sec the max vide length or we can get more? @dawn trout
Hi! I’m one of the people who started
No one actually. You have to send it here and staff will respone like they did just now
Has anyone figured out how to get the
Hi, I'm located in Wales, UK and was hoping Opal would have roleld out. I'm sure it was announced as being available in the UK? If not, anyone know when the rough roll out is?
Hi, I'm located in Wales, UK and was
I use opal all day for stock market anaylisis..(fantastic) It is to me a great prototype (mini app) next step is to get real -AI studio then Vertex AI ...
I struggle with after the api key..hit a wall.... So I go back to my opal stock apps and life is easy again....lol
Suggestion to the team if reading these comments.. Take my situation and make it super easy... perhaps a chat bot in vertex ai studio.
Our business hits the same wall ... Whats interesting is we think ourselves as tech savy ,(took the Gail course, great!) using workspace intesly in our insurance biz... But the WALL after the api key... lol I should make a veo 3 vid about the WALL..in my freetime ...lol ditto
During my Opal workflow, one step requires using Nano Banana Pro to generate an article cover. I've now exceeded my quota. How can I bypass this quota restriction? Are there any methods available?
During my Opal workflow, one step
Testing YouTube creation
I have a question. Since Opal is part of Google Labs, are there any development plans to integrate other Labs models so they can work within Opal? If such plans exist, could you provide a rough idea of which models will be included and approximately when?
I have been using Opal quite a lot lately and it has been very helpful. Thank you!
👍
I have a question. Since Opal is part of
I use opal all day for stock market
Question: Can’t get the “Save to Google Docs” step to include generated images, even though the Console shows that an image was generated -- am I missing something, is this a bug, or is it a limitation of Opal?
Question: Can’t get the “Save to Google
hello, how can i fix this issue? I have another prompt to check with user before moving the next flow. Right now the only way is the create 3 opal app?
hello, how can i fix this issue? I have
Yo Gemini 3 flash general availability a go go?
Seems a so, so.
Lemme at them utilities though
Still waiting for opal in Poland..
How can I fix this issue? What does this error mean? I've been trying to solve it for 30 minutes, but nothing is working😂
is opal avalibale in germany?
Hello, as a friendly reminder we cannot generate images or videos in the server! Please do not post your prompts in the channels as it will not generate. If asking for advice or tips on your prompts do so in the appropriate channels.
Hello, maybe wrong place, but can anyone help me to find information about when will be Opal availible in Czech Republic? I am really furious to test it! 🙂 thanks
Hi there, I’m exploring ways to connect the automation workflows I’ve built in Opal with the backend of my application. Is there a recommended approach for integrating them, similar to how n8n handles backend orchestration through APIs or webhooks?
Hello, I tried to find that information too, but I cannot. So I looks like we have to wait, sadly 🥲
Hello, we wanted to remind you of our server guidelines, stated in #start-here・📍 : “Currently, our team can support engagement in English; the discussion in this server will largely remain in English for now.” Please keep in mind that non-English discussions and content may be removed. Thank you!
Hi folks, bumped into Opal today and I loved the idea! Ran into some issues when prototyping the 1st app
- I can't directly edit the prompt in each step, this sucks because I see only small update is needed and I can't do that directly
- When using prompts to edit the prompt within a step, I always get error and it failed
- In my multi-step workflow, I can't see what's the output of the each of the step, making it hard to debug why the result is suboptimal
Can I get some helps on the above? Appreciate it!
when moving from one node to another, do I have to use only mouse clicks?
Hey @normal rampart! Welcome to Opal 👋
-
You can edit hte prompt in each step directly! If you're in GeminiApp, then click "open in advanced editor" (or go to opal.google directly). Once you see the graph, click a specific step and you'll see a side panel open up with the exact prompt. Click into the prompt to edit.
-
can you give me an example of an edit prompt that failed? Sorry it's not working for you!
-
In the graph view, open the side panel and click "console" while you're running your opal. You'll see all the interim outputs there!
Hi there, I’m exploring ways to connect
@elfin tulip are you able to tell if anyone has seen my bug report? https://discordapp.com/channels/1172568727942860810/1464802948373680188/1464802948373680188
The gem I made was working incredibly well, but lately I keep getting invalid blob ids and batch generations beyond 8 are becoming invalid
@Megan Li (Google Labs) are you able
Hi everyone, I'm facing some issues with the General generation node when trying to generate a specific number of images (e.g., a set of 7 images).
No matter how I optimize my prompt, the output quantity and model behavior seem very unstable and unpredictable. I'm observing the following inconsistent behaviors:
- Over-generation: Sometimes the node invokes the model/tool 7 separate times, but the total output exceeds 7 images (e.g., each call generates multiple images).
- Under-generation: Other times, with the exact same prompt, it only triggers a single model call, returning a batch of just 2-3 images instead of the requested 7.
- Non-deterministic: Even when I copy the exact same prompt into a brand new workflow, the execution logic changes randomly between the scenarios above.
My Goal:
I need to generate a precise number of images (e.g., exactly 7) to create a consistent set.
Question:
Is there a reliable way or specific configuration in the General node to strictly control the number of tool invocations or the batch size per generation? The prompt instructions alone seem insufficient to control this loop logic reliably.
Any advice or workarounds would be appreciated!
Is there a way to see the initial prompt i submitted to make the opal mini app?
come'on 🙁
It's been like 4-5 weeks
I wanna try this, I live in Belgium, EU
I have google AI pro account
comeon let me try out the new toys 😄
Is there a way to see the initial prompt
Thread
Hi, I have created a new email box, however I cannot use the mini app. It is said the location is problematic? Do I enter necessary a US address ? Or can I have access to the app. differently?
Is anyone facing any issues on Opal - I'm geting this error
Is anyone facing any issues on Opal - I'
Has been happening from last night
Tried again and now I'm getting a network error.
Looks like there is some latency at the network layer - timing out
It went past this screen. Still same issue as before. It's gone past one node - but failing on 3rd node.
It went past this screen. Still same
How do we utilize the Flow Control Redirectors?
Hey there, Google FTE here in REWS and working on an opal app. Not an engineer, but loving how opal makes it so easy. I’m wondering if there’s a way to have Opal use an API to connect data automatically from Plx. As of now I have it set up to manually upload .csv data sets but I would love it to pull the relevant data automatically. Any tips are appreciated!
Also if any Google FTE with more Opal experience would like to partner on this please lmk.
I hope there is a regular trigger start node. This can allow some like morning inspiration within like sending image via Gmail or drive
Thread
I hope there is a regular trigger start
Bandoli game khela
@bold sphinxHey, for your safety, please don’t post your email publicly. This server is public, which means anyone in the world who wants to join will be able to see what you wrote. That puts you at risk of scams and phishing attempts.
If you noticed a bug with limits, please describe it in #1238621511401734254 (without any personal information like email)
New to this and its looking relatively like n8n
man i should of been here months ago
and what are the reate limits or the limitations?
is coming gemini 3 flash with deep research?
Good day friends. I got seriously tired of the AI race and have seriously decided to just remain in Google's ecosystem.
I think we are in a time where EVERYTHING will be getting upgrades, so it is better to just embed into an ecosystem of tools and grow with the ecosystem.
I still use n8n to connect a lot of what I have going on, but I am learning about Opal.
Is there anyone here that uses n8n and can tell me there biggest differential experience with opal?
(i tried using it myself and even working with gemini to understand it a bit more, but I am still leaning towards n8n here --- I DO want to grow with opal if it is a viable replacement)
<@&1410743137319522395> QUOTA RESET REQUEST
Veo 3 Pro → Opal daily limit still exceeded (12h+ after ticket)
labs-bugs-feedback post submitted but no reset yet
Please reset quota ASAP! Thanks!
@Opal QUOTA RESET REQUEST
Hi guys, I'm new here and I'm building an app in Opal. I'd like to be able to have my app add events to Google Calendar, but Google Calendar doesn't appear in the Tools section. That's why, even though I enter my calendar ID, it obviously gives me an error. I'd like to know if you know how I can add Google Calendar functionality.
I have the following idea: I’m trying to build a PDF translation orchestration agent that runs in three automatic steps. First, it checks how many pages the PDF has. If the file has ≤ 30 pages, it sends the whole file directly to a translate node; if it has more than 30 pages, it uses a code tool to split the PDF into chunks of up to 30 pages, then sends each chunk to the translate node in sequence, looping until everything is processed. Finally, it concatenates all the translated chunks in the original page order and sends the result to an Output node.
However, I’m running into a weird issue: sometimes I don’t even connect the agent node to the Output and Translate nodes, but the flow still runs and works “kind of okay”, except it stops halfway. When I connect everything properly (agent → Output, agent → Translate), it doesn’t return any result at all.
I’d really appreciate any help or hints on what might be wrong here. Thanks a lot!
Despite repeated instructions not to include background music, background music continues to be generated, causing repeated failures when attempting to create a quiet ASMR video.
Additionally, even though this is an image-based video generation process, the system only references the provided image and generates newly redrawn visuals instead of preserving the original image as the basis of the video.
This issue did not occur until about a week ago, but it has started happening suddenly. Could you please explain what caused this change?
At this point, the concept itself seems to be interfering with the video generation process, making it difficult to produce the intended ASMR video.
Hi everyone. I'm new to Google Labs. I wanted to check it out and ask when Opal will be available in Germany and what it can do?
2969 PETABYTES/ second
Hi Guys, I keep getting error saying Opal has limited something try after some time why ? need to increase the quota ?
Think 👍
what you on the nasa's computer? lmao
Hello all, I'm new to this and trying to build an Opal agent for my fitness journey. This is the first time using Opal, but I can't make it work because the NL prompt is one-sided. I'm not getting feedback or meaningful error, so I'd appreciate any guidance to find and fix any issues.
Solution I am trying to build is a chatbit take the fitness program overview and food preferences and based on that the app should provide suggest based on the research.
Would it be necessary to use a Google Workspace organization account, or to perform any additional configuration, in order to use this service?
Alternatively, is Google Opal currently not available for use with Google Workspace organization accounts?
If organizational use is planned, could you please advise on when it is expected to become available for use within a Google Workspace organization?
Hello all, I'm new to this and trying to
Hi guys, I'm new here and I'm building
Tired of inflation eating away at your savings? Crypto is immune to the pitfalls of fiat currencies.
I have the following idea: I’m trying to
Hi
hey bro
Hi, I'm new here in the Opal discord section at Google Labs.
Does everyone have these problems with video creation and first frames?
For example...
Hello World
The initial frame is similar to the photo I give as a reference.
Hello
hopefully it gets patched before someone gets a cardiac arrest 🤞
i like the pfp mate
Please help me delete my pesonal file I have uploaded to Opal.google
Hello. I'm having a problem using Opal.google and I need your help. When I create a gem from the Gemini app, Google redirects me to opal.google. I created the gem and tested its functionality. During that process, I uploaded a personal file to the gem, and then deleted the gem because I no longer needed it. However, the file I uploaded is now on a public link https://opal.google/board/blobs/... that anyone can access. Can you help me remove and delete that uploaded file? Thank you very much.
Please support video generation in parallel. Because of rigorous cool down, It does not work.
Please help me delete my pesonal file I
hey i am Adii from india i am searching for aiml certification course, any suggestions?
Hi, I'm new here in the Opal discord
How to activate Experimental Features and why wasnt it announced when it begun. (It really deserves one)
I heard there was which go to a certain step or refer a step and instruct an Generate Block to go to the mentioned step
Thank you for the drop, Opal Team! I really liked the Gemini 3 flash in my app!
Guys, what kind of things do you use Opal for?
Hi everybody, why I can’t connect to opal from my mobile phone (iOS) and MacBook ( the opal is not available in your country yet) , but easily connecting via my computer ???
Hi everybody, why I can’t connect to
The issue has been resolved. The system now correctly generates new Google Docs and Sheets for each run, rather than appending content to existing files
The issue has been resolved. The system now correctly generates new Google Docs and Sheets for each run, rather than appending content to existing files
Looks Really Good
Thank you
No Problem. How ya doing? What are your hobbies other than Opal?
Lately, I have been feeling completely free from all my obligations. My only hobby now is simply drinking coffee as soon as I wake up.
I am also constantly looking for ways to improve my workflow using artificial intelligence. It is truly useful for many things; you just have to sit down and explore what you can do with it.
Main black oil
hi
I’ve documented my full requirements in the attached file, but in short: I’m trying to build a context‑aware fitness & nutrition coach that can read my Google Drive workout program, OCR my workout screenshots, track progress, log food/macros, use my bloodwork, and respond naturally while keeping long‑term context.
I’ve tried multiple approaches and experimented with the sample app shared here, but I’m still not able to get the full workflow working the way I need.
If anyone can guide me on how to structure this in Opal — especially around context management, Drive integration, and interactive chat behavior — I’d really appreciate it.
Thanks in advance for any pointers.
Hello everyone,
Do you have news when opal could to use in Europe using normal agents without VPN?
Hello everyone,
Trying out Opal for the first time. Why do you need to have access to ALL folders and files in my Google Drive to be able to use Opal?
Can't we do permissions for select folders and files only?
It was one of the most requested features after Branching and Looping. They might be working on it as part of the new update. (A New Update is coming soon. I am not staff)
Good morning .
Opal team members... Did you see my majestic thread about the video issue and feature suggestion?
It was
Can everyone explain to me how these things work and what they're used for?
Is "Use Memory" new?... I saw "Go to" but I don't know what it's for.... Is it to go to a specific webpage?
how can use opal like this???
I am not sure I fully understood your question, so feel free to rephrase it. That image is a mockup showing the future of Opal. It is a conceptual preview of upcoming features inspired by platforms like Make and n8n. While those functions are not available yet, they might be possible in the next update with the introduction of agent mode. Essentially, the image represents what Opal is expected to become.
That’s a great idea. Let’s start using APIs and external platforms in Opal too! 😎
I keep getting issues with OPAL quota . . why
Please help me understand as well
I keep getting issues with OPAL quota
That’s a great idea. Let’s start using
The images display Opal's automation tools, including 'Plan and Execute' where Gemini 2.5 breaks down complex tasks, 'Code Execution' for running real-time scripts, 'Go to' for rapid navigation, and 'Use Memory' to recall user context
I have an idea for an Opal feature. I sent you a Gemini video so you can understand it better.
The idea is to create an agent or assistant focused on document editing. The process would be as follows: you send your document, select the agent responsible for the editing, and set a timer.
The assistant would be activated periodically, analyze the document, and make the necessary corrections continuously.
This is somewhat possible with Opal. Although 2 things are very necessary.
-
We need to use Plan & Execute Model (Available in the Generate Step Menu) to conduct the changes on a Google Docs File added as an Asset (This is necessary if all notes are combined as docs, which is necessary if you had dozens of notes and each note is its own file)
-
The Ability to swap assets without removing the connections
In this special episode of The Neuron Podcast, we get exclusive demos of three cutting-edge AI tools straight from Google Labs with the product managers who built them.
Featured Tools:
• 🎨 Mixboard (Jaclyn Konzelman) - AI-powered concepting board that transforms your ideas into visual presentations using Nano Banana Pro
• 🎬 Flow (T...
Would Jules be a good option for that on schedule? Perhaps having Jules call an Opal on schedule?
It sounds like an interesting alternative, thanks.
(Opal short sleeve) Outdoor photography app
https://opal.google/app/1L7H4W0AYVaaAN03neeQ0m_TYkYsdDH-e?shared
Its a good idea. I could see it nice to have a regular "editor agent" to look over a book, essay, article, etc. on a routine schedule(hourly, daily, weekly) and give feedback.
Hey there...i was planning to go to a hackathon with no code thing but... The problem is backend can anyone suggest which tool should i use with like antigravity or cursor...or to get working backend with ai..
creating an opal!! 😄
It has been a few days since the last tease of the "Agent Mode" Update. Are we getting on 10th Feb to go hand in hand with update schedule of Opal where updates will come on Wednesday's?
How do I use "use Memory" tool?
.
When is Google Opal coming to Sweden?
Yes how to use MEMORY, could not find any document as well for that ?
Has anyone tried using Opal for market research or focus groups?
Hope opal could save as json like n8n. that way, I can edit with lots of ai
know it's easy to realize. wish come very very soon
What scale are you thinking-quick prototype or full enterprise rollout?
can opal help me make an email sender bot?
can opal help me make an email sender
Hope opal could save as json like n8n.
Hello, I am pretty new to AI in general and came upon Opal as I was tasked with building a custom GPT through my AI grant writing course. I am building an Opal app that can generate grant narrative skeletons, full narratives and grant applications. However when I input my information into Gemini, it just spins and appears stuck. Screenshots here. Does anyone have any suggestions?
Why opal is showing this?
weird, can you send the link to your opal? We'll take a look!
Hello, I am pretty new to AI in general
Show editor section
Hello!!
seems to work for me. Are you still getting the error
Open the algorithm in editor, check out which box is highlighted when you run your app in preview mode resulting in this pop up. Most likely you have to select that box and activate web search tool. It is easier to work on a bigger screen (not mobile)
Hello, i am trying to save an Output in a specific Google Drive Path. Is that possible?
You can save output in gmail drive
Whoops

How?
Usually I export it as html file
Greetings, Is the Agent Mode coming this week or next week or somewhere before the end of the month? Take your time, No Rush. Just wanna know
imminent, but not exactly sure when
what is "use memory"? and how to use it? what is agent mode?
How do you use the new Go To Tool? When I use it it skips the route ahead and outputs the current response early. Additionally there's a use memory tool, how is that utilized?
Here's where I tried to utilize it, but it failed and skipped both paths and outputs and just outputted the check for cat box. I thought maybe it could be used as a if/or or a subagent system, but no luck getting this tool to work so far. Worked for anyone else?
I'm working on something for a client, and after trying-out Opal (and LOVING it), I'm hoping I can use Opal instead of Adobe Express for this content creation I'm working on. Creating reels using CSV import on adobe express is a pain.
I have about 50 reels to create. Each one of them with about 5 images (moving ken burns style) each. I already have the images, and text overlay. Could opal handle this for me? Asking cuz I'm seeing a few limitations. I might use a vibe coding tool for this instead though
Got it!
I'm working on something for a client,
Am coming up with a meme coin I need a marketing manager that can help me hold half of my supply I will pay $400
Hey everyone 🤗
guys, do opal have public api or something like that yet just like other tools out there?
Hey team, sorry I've been quiet lately. Thanks for the updates, like Dimitri on agent mode.
Quick idea: Let's add a "Keep Us in the Loop" channel just for Opal/Google folks. They could post progress there, like daily blogs on what they are working on, such as agent mode. That would be a big help.
On agent mode: What is it like? Is it similar to the Gemini 3.0 Flash agent site? Can it handle web tasks, interact with sites, or more advanced stuff like integrating with WhatsApp, Telegram, or Instagram without complex setups?
For videos: Fixing creation tweaks sounds great. This could enable stitching 8-second clips for longer ones and handling frames. Maybe external tools could be used for a consistent voice.
Veo questions: Is there any chance of adding a duration selector (4, 6, or 8 seconds) in Opal, like other sites have? What about future models like Veo 3.2, 3.5, or 4 around Google I/O, or perhaps sooner due to Seedance 2.0?
Also, why not reward heavy users with bigger limits via streaks? Keep it generous for experimenting—no premium mode, please! Let us play and mess up without needing credits. (Or perhaps the most logical thing would be large free limits, and much higher limits if you have a Google plan. )
What are you working on next? Updates on timelines would be awesome.
See you soon!
@quaint eagle, Greetings, If conditional is going to be on Opal? Can you take some inspiration from Scratch (ik who uses it nowadays. I feel like, It's the most visually easiest thing to understand (a lot easier for causal opal users to understand and modify things). I am a Vocational Desktop Publishing Student and Learning Communication Design on my spare time
Also, Ability to add boxes which we can enter prompts or descriptions (it should be able to be selected (to modify via AI) and it can be removed from prompt or modified easily without affecting other sections of the prompt, It should be able to lock out. The Users/AI cant modify this section, only the creator and who else they allow them to (if permissions can be shared).
hi everyone. i get this error repeatedly, am i doing something wrong?
wondering if it's related to model selection. do you have to have a pro account to use pro models?
Hi, that usually happens because that model has reached its usage limit.
I think I also forgot to mention that, according to the last time I checked, Opal was missing Gemini 3 models for HTML rendering.
For example, if you were using the Gemini 3 Flash model, using it repeatedly you reach your daily limit, and then you could switch to another model like the Gemini 2.5 Pro.
I'm having an issue where failed video generations—due to policy triggers—are still counting against my daily limit, even though I haven't successfully created anything.
Hi everyone, i’ve been using Opal to generate videos for a couple of months. For some reason, I can’t generate ANY videos at all anymore. I went on today to generate a few, and immediately I wasn’t able to. I haven’t generated any videos in about a week so I know my quota isn’t exceeded. I tried with 4 different gmail accounts as well, same issue across all of them. I keep getting random error messages as well. The last one said “Load Failed”, the one before that said “Failed to fetch” the one before that said “Service Unavailable”. I checked in google devtools and found under the network tab that “executeStep” keeps failing with the status code “503 Service Unavailable”. Anyone else having a similar issue?
I was in the same situation... I'm going to try...
mine is working now! Not sure why, but I did try to fix it for like two hours lol
Love this! We'll definitely do more to keep you all posted on what we're working on. Smaller things that are in flight now: NotebookLM integration for grounding on a notebook, Mobile optimization (you can create Opals on mobile now, still one last iOS bug but WIP), new sharing flow so you can hide your promtps/graph view when you share with others, better multi-modal UI options for inputs, and more.
As Dimitri mentioned, the biggest step change is the ability to have an agent step. This is coming very soon and will unlock looping, conditional logic and memory. Agent steps will be able to choose tools and models to call all within one step, you just need to add your prompt as your objective. We don't have 3P integrations yet, but we're investigating how to unblock this on our side. If you have specific partner integrations you're most interested in, please let us know!
That's encouraging news, Megan. Thanks to the entire team for paying such close attention to all the community's ideas. We're really looking forward to this update.
We'll always be here to help improve the app with feedback, questions, and anything else needed. Especially now, given what you mentioned. We truly appreciate your transparency and interaction with the community.
I’m in the Tampa Bay Area of Florida. Is mixboard available here?
Hello, it's supposed to be yes.
It’s available now, at least for me. I logged into Opal and noticed the 'last updated' date had changed (though not to today's date). I poked around and found the 'Copy JSON' option in the three-dot menu inside the flow panel.
hallo
hi
@magic saffronAs stated in #start-here・📍; currently, our team can support engagement in English; the discussion in this server will largely remain in English for now.
hi
If you want to create videos with Opal using the simplest workflow possible, you should start by adding a User Input node and a Generate node set to 'VEO' Connect the User Input to the Video Generator (VEO) and make sure the input node is configured to send both text and images. Finally, select the specific video model you need; that’s where you can decide whether you want a vertical or horizontal video using Google Veo 3.1.
Keep in mind that when using reference images, you might run into consistency issues between your input and the generated video. To stay informed, make sure to follow this Opal channel, where the Google team shares the latest updates and features. If you encounter any problems, you can report them here or via the feedback section in Opal. The team mentioned that new things are coming soon, so hopefully, a fix for that video issue is on the way.
If you have any other questions, no matter how simple they may be, please don't hesitate to ask.
Hello, I am a developer from South Korea.
In order to properly verify whether this program works well, it seems that I would need a very large amount of tokens. I am actually willing to pay to evaluate the completeness and performance of the program, but I cannot seem to find a clear way to do so. I have been trying to push forward, but honestly, I am currently exhausted.
I have three questions:
Is the generation limit reset once per day, or how does the reset cycle work?
When will this service become fully paid?
If immediate full monetization is difficult, is there any way for me to pay a certain amount now and run experiments in a more flexible or expanded environment?
why isnt opal working for me? it says not available in my country, but my country is on the list of available countries
same for aistudio btw
knock knock~
Hi @sullen rune , I think the people at Opal will be able to answer you tomorrow, and with more precision than the rest of us here.
Thanks, I believe I can become a key contributor to the Opal program.
I’d love to collaborate and help it grow further.
Currently, I am independently researching it in Korea.
Wow, that’s great to hear! It really is such a versatile tool. It fits perfectly into so many different workflows—from creative content creation to deep research—making it just as valuable for independent creators as it is for corporate teams.
Also, the team was just mentioning here that they’re working on some really exciting features. They’re definitely going to take our workflows to the next level.
Hi, would be Opal user here in the UK, unfortunately we appear to be ecluded by Google for the the moment. Are there any UK based members that have managed o get a work around in order start?
VPN?
Yes, but I need my Google ID to sign and they know I'm based in the UK. Are you suggesting a new Google account?
Hello, where does the credit for the images and video generation used in Opal come from?
consider separated with other services
📅 hi y'all. Today is a holiday in the US. Chat with you tomorrow
Okay, Thanks for informing us, Dimitri
Okay, thanks for letting us know.
So @sullen rune, you're supposed to get a response tomorrow.
Hi All!
Im trying to get access to Opal to start testing it but I can't do it even with VPN located in US. Some recommendation?
Hey @quaint eagle
I received a suspicious message in my DMs from a user who came from here. They asked me to create an account on a freelance platform and give them access so they could create contracts. And he would pay me $100 for successful signings.
I need support to report this practice.
that's a task for <@&1366831844502274180>. I build Opal 😄 I don't moderate Discord
Hi feel free to open a Modmail by messaging @versed lotus and we can help you out!
Ooh, thanks!
Sure sivv, thanks!
guys, do opal have public api or
Hi @ashen flume
hi
Hey team, sorry I've been quiet lately.
I'm having an issue where failed video
Ji
😍 🚀