I click on the invoke_starter.exe file, extract the files, then after the extraction begins I get "an unexpected error is keeping you from copying the file...error 0x80004005 unspecified error, and it lists
commandline.bat:
type: windows batch file
Date modified: 5-6-23 2:16 pm
size 843 bytes
I click skip, but then many of the files after this have the same problem. "
I created a folder on the desktop, and tried to extract it there. Then I created a folder on the c drive just called invokeai, and tried to extract it there and got the same error message.
Edit: Can you install 7zip and unzip it with that (or WinRar, etc)? Apparently windows doesn't support the zip archive type that we used. :/ Sorry for the trouble.
I downloaded and launched, everything seems to be fine and there are no problems in the settings, but it generates just pastel-colored squares instead of a request. What could this be related to?
can anyone help with my problem ? i keep getting this error...what does it mean and how to solve it.
OSError: [WinError 193] %1 is not a valid Win32 application. Error loading "D:\invoke ai\invokeai\invokeai_standalone\env\lib\site-packages\torch\lib\cudnn_adv_infer64_8.dll" or one of its dependencies.
It appears that you are on a 32bit system (e.g. old Win7). The starter itself only supports 64 bit.
You should still get it to run with those steps: 1) Open invokeai_standalone/invokeai/invokai in a text editor (e.g. "Editor") 2) In the lines --outdir="" etc add the paths that you want to use. E.g. --outdir="E:/invokeai_standalone/outputs/" 3) You can disable the nsfw filter by removing the last line (--nsfw_checker) 4) Save the file 5) Double click invokeai_standalone/helper.bat
just download invokeai_standalone.zip (8.2 GB) via itch, It was fast and it worked (run) without any problem. must say this was very early (1 to 3 am) so maybe it helped about Installing process stucks (happened to me too the first time) I'll see how it works tomorrow! :)
While it is listed as standalone, does it still need the installation script that downloads dependencies such as python, as compared to older releases such as aiimage where it was a simple extract any it was immediately ready to launch and use (fully self contained) with no additional setup process?
The installer from Itch app is still freezing after downloading Invoke AI.
Installing process stucks at about 10%. I've waited hours to finish and no change.
I've deleted, redownloaded and reinstaling 3 times.. still freezing
From a user perspective, nothing much should change. You can download it here directly or from the itch app (thought the app doesn't have an advantage anymore, since it cannot create patches for our external downloads. :( ).
when I was using it the tool was pretty good but today I tried to update it and now it won't work I tried to re-download it from here but it says error what I liked this and people in Discord dont seem to see my messages so they didn't help
it seems to be having an error, or its owner is changing the version, I also downloaded it last night but the speed is very slow, this morning it is no longer available
Hi, I realise the VAEs are not getting correctly converted when converting a checkpoint/safetensor model to diffusers. The VAE does not seem to take effect and the final rendered image will appear to be washed out.
Thanks for the report! :) There were some patches since then, so the problem might be solved. If not, definitely poke the InvokeAI team about it on their discord. <3
3.5x speed upgrade with Nvidia .dll replacement working for InvokeAI Standalone 3.2.0 and a 2x speed increase for standalone 2.3.2 and the regular install of 2.3.4.
I've replaced the CUDA files for invokeAI 2.3.0 and it makes it 3.5x faster with my Nvidia 4090. I'm running InvokeAI 2.3.0 at 28 it/s just by replacing the CUDA .dll files alone. However, I cant seem to get the latest to be as fast by replacing the .dll files but at least it's still 2x faster on InvokeAI 2.3.2 and also the normal install of the latest 2.3.4 with .dll replacement. Not sure why I cant get it going as fast as 2.3.0 at 3.5x the speed. For in-painting I'm still using Invoke 2.3.0 with the upgraded .dll's and loving the 3.5x speed. Any idea why the versions after 2.3.0 are not performing as fast as InvokeAI 2.3.0 with the latest CUDA files? anyone know how to get more speed out of Invoke 2.3.2-2.3.4?
I Downloaded the Nvidia zip file here and used the .dll files found in the bin folder:
I figured it out. The improvement you are able to see is because nvidia fixed the driver compatibility issue with CUDA 11.8 on the 4k series of cards. Everyone using a previous generation is already running at maximum performance.
So far my experience has been amazing. Very user friendly and fun to use! Are the use of LoRAs or Dreambooth style training something you are considering?
You can run the update_experimental.bat, then you get Lora support. :) Or, if you installed via the itch app, you can also wait some days, then I get the patch ready.
I have it installed through the itch app. Would it mess things up if I update through the .bat file? Approximately when do you think you'd have it updated on itch?
The release has been updated. Make sure you run "update_experimental.bat" from your invokeai folder.
Chose option 1 and press enter. The update process will begin.After not too long your will receive a message informing you the update was successful.
You may now launch InvokeAi using the launcher (do not pay attention if the launcher says version 2.3.2, it's just a bug). What matters is the version displayed in the Web UI, in the upper left corner.Do not forget to post your reviews, questions and your own creations here.
For models please visit Civitai. Feel free to start using InvokeAI with a powerful model:
did the errors I encounter are corrected now? :3 , should try to reinstall? I dont care doing a few steps before start generate images tho. Im glad this works cz wasn't able to install via github but whatever xD
I was able to update via -update_experimental.bat- now, a few days ago this wasn't posible, when u try to start after installation, on the console it will appear an error, ("tuple index out of range") and ask u if u want to (re)install the models, just type n for cancel that, it will open the page, however, there's a little "bug" with the token "thing" i think, u have to load a ckpt using model manager, then click on update model, that solve the issue, then change the sampler option, lets say ur using dpm++ sde karras, change for whatever option, generate an image and then go back to the one you use, if not the image will be generate really bad xD, sorry if this are broken english, isn't my main languaje :3
Hello, thank you very much for your comment, I am oriented to update this program, I had been looking for days how to do it, also for the clarifications apart to correct it, by the way my original language is not English either so we are the same XD.
sometimes this works wonderfully for me, while others i can't seem to get it to stop generating empty black images and i wish i understood why. using a 2080 super. thanks!
Sometimes it depends on the checkpoints you have installed. I recommend updating to version 2.3.4 since it is also faster. You should convert checkpoint to diffusers for maximum speed, it literally halves the rendering time.
i Use Itch to install Invoke AI just because it says i can easily upgrade InvokeAI. but now 2.3.4 is out, when I click "update" here , nothing happened. it said , nothing new. ...
That's because this is a standalone version "by Sunija". So I guess he will update it when he can. You can download the other one as well, I ran both for a while.
You can update with the "update_experimental.bat" in the standalone folder. :) But I'll also upload the update tomorrow (and integrate a button for the update script).
I am newbie, i use Itch just because it says we can update InvokeAI as easy as like upgrade a game (when new version is out) , however now like you said 2.3.4 is out, but I click "update" button here, nothing new.
or maybe wait a couple of days? or something I did wrong? thank you.
The current version of mine installed via Itch is 2.3.2 ...
Hm... that looks all fine. You could try to reinstall it (make a backup of your output folder first!) or poke me on Discord (Sunija#6598) and we solve it together. :)
It cannot run on phones, because they don't have the required performance. :( You can run it on your PC and then access it on your phone via the link in the starter.
1) Run the starter as you normally would. 2) Check the "Share access" option on the left side. 3) Restart the starter. 4) Open the link next to "Your Wi-Fi:" on your phone.
Sadly, I don't know. :( I hoped they would be available months ago. Apparently there are some less fancy UIs that already support AMD cards. You could google about those (or somebody posts them as an answer here).
File "E:\Itch\Invoke_Ai\invokeai\invokeai_2_3_0_standalone\env\lib\site-packages\invokeai\backend\modules\parameters.py", line 45, in parameters_to_command
if "init_img" in params and len(params["init_img"]) > 0:
TypeError: object of type 'Image' has no len()"
Suffice to say I have put in an image into the prompt and also tested various settings, result is the same seemingly regardless of model used.
Of note, Unified canvas still works, so it can still generate based on another image, just something going wrong with img 2 img specificly.
← Return to tool
Comments
Log in with itch.io to leave a comment.
Yes, it includes LoRA.
When you got the newest version, there is an "open lora folder" button in the starter. Click it, and drop your loras there.
In InvokeAI, there's an "Add Lora" button below your prompt, which you can click to add loras. :)
I click on the invoke_starter.exe file, extract the files, then after the extraction begins I get "an unexpected error is keeping you from copying the file...error 0x80004005 unspecified error, and it lists
commandline.bat:
type: windows batch file
Date modified: 5-6-23 2:16 pm
size 843 bytes
I click skip, but then many of the files after this have the same problem. "
Any fixes for this?
Thanks for reaching out! :)
In which folder did you try to extract the zip?
(There are some folders, e.g. system files, etc, where windows won't allow the extraction to go through.)
I also got the same error while opening the zip file
I created a folder on the desktop, and tried to extract it there. Then I created a folder on the c drive just called invokeai, and tried to extract it there and got the same error message.
Edit: Can you install 7zip and unzip it with that (or WinRar, etc)?
Apparently windows doesn't support the zip archive type that we used. :/ Sorry for the trouble.
I downloaded and launched, everything seems to be fine and there are no problems in the settings, but it generates just pastel-colored squares instead of a request. What could this be related to?
Thank you for reaching out! :)
Can you send me a screenshot of your outputs on discord (Sunija#6598)?
Yes! I would like to send you a screenshot of my problem, I sent you a friend request in discord
I tried via itch and it says hosted on a 3rd party incompatible website. Can't download via itch
I had the same error yesterday. Seems to be gone now, try it again.
I added the mirror download link again. :)
Thanks for the notice that it's working again! <3
Awesome.
Hi!
What version of invokeai is this?
The most recent. :) (2.3.5)
hi guys
can anyone help with my problem ? i keep getting this error...what does it mean and how to solve it.
OSError: [WinError 193] %1 is not a valid Win32 application. Error loading "D:\invoke ai\invokeai\invokeai_standalone\env\lib\site-packages\torch\lib\cudnn_adv_infer64_8.dll" or one of its dependencies.
Press any key to continue . . .
It appears that you are on a 32bit system (e.g. old Win7). The starter itself only supports 64 bit.
You should still get it to run with those steps:
1) Open invokeai_standalone/invokeai/invokai in a text editor (e.g. "Editor")
2) In the lines --outdir="" etc add the paths that you want to use. E.g. --outdir="E:/invokeai_standalone/outputs/"
3) You can disable the nsfw filter by removing the last line (--nsfw_checker)
4) Save the file
5) Double click invokeai_standalone/helper.bat
Its strange though because I'm on a 64bit windows 10 pc. ill try installing it again :)
just download invokeai_standalone.zip (8.2 GB) via itch, It was fast and it worked (run) without any problem. must say this was very early (1 to 3 am) so maybe it helped about Installing process stucks (happened to me too the first time) I'll see how it works tomorrow! :)
While it is listed as standalone, does it still need the installation script that downloads dependencies such as python, as compared to older releases such as aiimage where it was a simple extract any it was immediately ready to launch and use (fully self contained) with no additional setup process?
It works exactly like aiimages, so there is no setup. :)
Only exception: If you do not have a nvidia card, it will download the 32bit models, because it cannot use the 16bit models to run on your CPU.
Cannot download with the itch app...says queuing...but for how long?
Download should be available again. :)
The installer from Itch app is still freezing after downloading Invoke AI. Installing process stucks at about 10%. I've waited hours to finish and no change.
I've deleted, redownloaded and reinstaling 3 times.. still freezing
You should download directly from the page instead... at least that works for me.
thank you, i'll try it again now that a'm at the ofice.
I'll let you know the resut.
It worked ! thank you !
I cannot download it with the itch app :(
I get this error : TypeError : Cannot Read property 'build' of undefined
In the InvokeAI Discord they said
"Itch cannot host invoke anymore, so the download won't work until we set up an alternative. :( Sorry for the trouble."
Download is back again! :)
From a user perspective, nothing much should change. You can download it here directly or from the itch app (thought the app doesn't have an advantage anymore, since it cannot create patches for our external downloads. :( ).
Hello Sunija! I am having difficulties downloading the invokeai? Is this a wrong on my end or are you working on something? Cheers!
Download should be available again! :)
Thank you for the report.
Any ETA on when the new install will be ready through the app?
Download should be available again! :)
when I was using it the tool was pretty good but today I tried to update it and now it won't work I tried to re-download it from here but it says error what I liked this and people in Discord dont seem to see my messages so they didn't help
Download should be available again! :)
where is the one click installer, from itch i get a error build not define
it seems to be having an error, or its owner is changing the version, I also downloaded it last night but the speed is very slow, this morning it is no longer available
Sorry for the trouble, the download wasn't available for a while.
Now it's back again. :)
Can someone help me with this issue please ?
Can someone help me with this issue please ?
after I click yes it doesn't work and I keep ending up here 1st pic
when click no it loads up and looks like its working but I only get this error when I try to generate an image pic 2
AMD 7900X CPU, AMD 7900XT 20GB GPU
I did the whole install over using the Itchio app now its working fine but its using my CPU instead of my GPU can someone help I'm a novice
Thanks for reaching out!
AIs currently only work with Nvidia card on windows. You got an AMD card, so it cannot use that for now. :(
Apparently AMD is trying to get AIs running, but it might take some months.
Hi, I realise the VAEs are not getting correctly converted when converting a checkpoint/safetensor model to diffusers. The VAE does not seem to take effect and the final rendered image will appear to be washed out.
Thanks for the report! :)
There were some patches since then, so the problem might be solved. If not, definitely poke the InvokeAI team about it on their discord. <3
Can we update to 2.3.4 ?
just using the "update_experimental.bat"
Yep, thank you for mentioning that! :)
Also, the new download is directly 2.3.5 now.
AMD announced ROCm will be coming to windows. Will support be added?
I think InvokeAI will definitely add it once ROCm properly works on windows. :)
3.5x speed upgrade with Nvidia .dll replacement working for InvokeAI Standalone 3.2.0 and a 2x speed increase for standalone 2.3.2 and the regular install of 2.3.4.
I've replaced the CUDA files for invokeAI 2.3.0 and it makes it 3.5x faster with my Nvidia 4090. I'm running InvokeAI 2.3.0 at 28 it/s just by replacing the CUDA .dll files alone. However, I cant seem to get the latest to be as fast by replacing the .dll files but at least it's still 2x faster on InvokeAI 2.3.2 and also the normal install of the latest 2.3.4 with .dll replacement. Not sure why I cant get it going as fast as 2.3.0 at 3.5x the speed. For in-painting I'm still using Invoke 2.3.0 with the upgraded .dll's and loving the 3.5x speed. Any idea why the versions after 2.3.0 are not performing as fast as InvokeAI 2.3.0 with the latest CUDA files? anyone know how to get more speed out of Invoke 2.3.2-2.3.4?
I Downloaded the Nvidia zip file here and used the .dll files found in the bin folder:
https://developer.download.nvidia.com/compute/redist/cudnn/v8.7.0/local_installers/11.8/
Then I replace the old .dll files with the new ones into the following: InvokeAI\venv\Lib\site-packages\torch\lib
I'm testing this for sure. Where did you find version 2.3.0
The improvement i'm seeing is by 1.5 seconds. Sometimes slower depending on the CUDA dlls version.
I figured it out. The improvement you are able to see is because nvidia fixed the driver compatibility issue with CUDA 11.8 on the 4k series of cards. Everyone using a previous generation is already running at maximum performance.
So far my experience has been amazing. Very user friendly and fun to use! Are the use of LoRAs or Dreambooth style training something you are considering?
You can run the update_experimental.bat, then you get Lora support. :) Or, if you installed via the itch app, you can also wait some days, then I get the patch ready.
Amazing work! I'll wait for the update on Itch. :D
I have it installed through the itch app. Would it mess things up if I update through the .bat file? Approximately when do you think you'd have it updated on itch?
Sweet, any eta? Wishing you the best :)
How can I install the patch via itchi app? is there a button? Help plz!
You can double-click update_experimental.bat in the invokeai folder. :)
its not working and I did all of the video, takes like half hour and still no 1 pic, I have to sleep.
I'm not sure I understand. What's not working ? The software or the update ? Join me on discord, I'd be glad to help.
The release has been updated. Make sure you run "update_experimental.bat" from your invokeai folder.
Chose option 1 and press enter. The update process will begin.After not too long your will receive a message informing you the update was successful.
You may now launch InvokeAi using the launcher (do not pay attention if the launcher says version 2.3.2, it's just a bug). What matters is the version displayed in the Web UI, in the upper left corner.Do not forget to post your reviews, questions and your own creations here.
For models please visit Civitai. Feel free to start using InvokeAI with a powerful model:
Crystal-Clear_v1 | Stable Diffusion Checkpoint | Civitai
did the errors I encounter are corrected now? :3 , should try to reinstall? I dont care doing a few steps before start generate images tho. Im glad this works cz wasn't able to install via github but whatever xD
I was able to update via -update_experimental.bat- now, a few days ago this wasn't posible, when u try to start after installation, on the console it will appear an error, ("tuple index out of range") and ask u if u want to (re)install the models, just type n for cancel that, it will open the page, however, there's a little "bug" with the token "thing" i think, u have to load a ckpt using model manager, then click on update model, that solve the issue, then change the sampler option, lets say ur using dpm++ sde karras, change for whatever option, generate an image and then go back to the one you use, if not the image will be generate really bad xD, sorry if this are broken english, isn't my main languaje :3
Hello, thank you very much for your comment, I am oriented to update this program, I had been looking for days how to do it, also for the clarifications apart to correct it, by the way my original language is not English either so we are the same XD.
sometimes this works wonderfully for me, while others i can't seem to get it to stop generating empty black images and i wish i understood why. using a 2080 super. thanks!
Sometimes it depends on the checkpoints you have installed. I recommend updating to version 2.3.4 since it is also faster. You should convert checkpoint to diffusers for maximum speed, it literally halves the rendering time.
i Use Itch to install Invoke AI just because it says i can easily upgrade InvokeAI. but now 2.3.4 is out, when I click "update" here , nothing happened. it said , nothing new. ...
That's because this is a standalone version "by Sunija". So I guess he will update it when he can. You can download the other one as well, I ran both for a while.
tHANKS
You can update with the "update_experimental.bat" in the standalone folder. :) But I'll also upload the update tomorrow (and integrate a button for the update script).
Update 2.3.4 is out !! Now with LORA and LYCORIS support.
Get it here: Release InvokeAI Version 2.3.4 - A Stable Diffusion Toolkit · invoke-ai/InvokeAI (github.com)
Thanks WarAnakin,
I am newbie, i use Itch just because it says we can update InvokeAI as easy as like upgrade a game (when new version is out) , however now like you said 2.3.4 is out, but I click "update" button here, nothing new.
or maybe wait a couple of days? or something I did wrong? thank you.
The current version of mine installed via Itch is 2.3.2 ...
Hey, great tool, by im curious is it possible to generate image with transparent background? Like UI elements, icons etc..
I don't think so. But you can try to generate something with an even background and then use another ai to cut out the foreground. :)
Any idea when the launcher will be updated via itch?
Tomorrow, I hope. :)
But you can already update today by running the update_experimental.bat in the invokeai_standalone folder.
Any chance the ControlNet or OpenPose add-ons?
In the future, yes. :) Version 3.0 will support plugins, which should allow stuff like ControlNet.
I was using perfectly, but stopped works and now keepping saying "System config requested"
What should i do?
Can you send me a screenshot of the error on Discord (Sunija#6598)? :3
unified canvas does not works anymore
Hm... that looks all fine.
You could try to reinstall it (make a backup of your output folder first!) or poke me on Discord (Sunija#6598) and we solve it together. :)
what is DPM++ SDE Karras and which sampler in the list is it?
I'm confused of the names
it's called k_dpmpp_2
"k" means Karras and "pp" means plus plus
Version 2.3.3 is out, change notes: Release InvokeAI Version 2.3.3 - A Stable Diffusion Toolkit · invoke-ai/InvokeAI (github.com)
Its not updating via itch. Any idea why?
Probably Sunija is working on his launcher.
Will there be an apk version
It cannot run on phones, because they don't have the required performance. :(
You can run it on your PC and then access it on your phone via the link in the starter.
can I go into more detail?
1) Run the starter as you normally would.
2) Check the "Share access" option on the left side.
3) Restart the starter.
4) Open the link next to "Your Wi-Fi:" on your phone.
Does that explain it? :)
Hey, is there any API to integrate it in your own code
The next big upgrade will bring a node system that should make it really easy to add your own code (as nodes).
hey, there is any prevision of when the support for amd cards gonna be up ? ;-;
Sadly, I don't know. :(
I hoped they would be available months ago. Apparently there are some less fancy UIs that already support AMD cards. You could google about those (or somebody posts them as an answer here).
Hi, img 2 img doesn't work, always get an error up by the model, and it fails. Command prompt has this at the end each time.
"command = parameters_to_command(generation_parameters)
File "E:\Itch\Invoke_Ai\invokeai\invokeai_2_3_0_standalone\env\lib\site-packages\invokeai\backend\modules\parameters.py", line 45, in parameters_to_command
if "init_img" in params and len(params["init_img"]) > 0:
TypeError: object of type 'Image' has no len()"
Suffice to say I have put in an image into the prompt and also tested various settings, result is the same seemingly regardless of model used.
Of note, Unified canvas still works, so it can still generate based on another image, just something going wrong with img 2 img specificly.
Any chance the ControlNet or OpenPose add-ons?