Hi @sand aspen. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
#Is it possible to use GPU for ESRGAN upscaling?
reef lanternBOT
Yes, it is possible to use GPU for ESRGAN upscaling in Invoke AI. The GPU usage is determined by the configuration settings in the invokeai.yaml file. The always_use_cpu setting should be set to false to enable GPU usage.
However, there have been reported issues on GitHub where users have experienced the system defaulting to CPU instead of GPU. In some cases, rebooting the system or reinstalling InvokeAI has resolved the issue.
If you're still experiencing problems, you might want to check the GitHub issues for potential solutions or report a new issue.
Sources:
- GitHub Issue 1763 (https://github.com/invoke-ai/InvokeAI/issues/1763)
- GitHub Issue 4197 (https://github.com/invoke-ai/InvokeAI/issues/4197)
- GitHub Issue 4202 (https://github.com/invoke-ai/InvokeAI/issues/4202)
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
sand aspen
@reef lantern Can I put this setting anywhere in the YAML file?
reef lanternBOT
Let me check the knowledge base for you. One moment! :books:
The always_use_cpu setting should be placed under the Memory/Performance section in the invokeai.yaml file. Here's an example of how it should look:
InvokeAI:
...
Memory/Performance:
always_use_cpu: false
...
...
Please ensure that the indentation is maintained as YAML files are sensitive to it. Any syntax errors can prevent InvokeAI from launching.
Source: InvokeAI Configuration (https://invoke-ai.github.io/InvokeAI/features/CONFIGURATION/#reading-the-invokeai-configuration-file)
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
sand aspen
@reef lantern This did not work. ESRGAN upscaling is still running on CPU
reef lanternBOT
Let me check the knowledge base for you. One moment! :books:
I'm sorry to hear that you're still experiencing issues. Based on the information provided in the extracts, it seems that some users have encountered similar problems where InvokeAI defaults to using the CPU instead of the GPU, even when the always_use_cpu setting is set to false.
Here are a few suggestions that have helped others:
-
Reboot your system: Some users have reported that simply rebooting their system resolved the issue (source (https://github.com/invoke-ai/InvokeAI/issues/4202)).
-
Reinstall InvokeAI: If the problem persists, you might want to try reinstalling InvokeAI. Make sure to delete the InvokeAI directory before reinstalling (source (https://github.com/invoke-ai/InvokeAI/issues/4202)).
-
Check for software conflicts: There could be conflicting software that's causing the issue. If you suspect this might be the case, you could try installing InvokeAI on a virtual machine to see if the problem persists (source (https://github.com/invoke-ai/InvokeAI/issues/4202)).
-
Report the issue on GitHub: If none of the above solutions work, it would be helpful to report the issue on the InvokeAI GitHub page (https://github.com/invoke-ai/InvokeAI/issues). This will alert the developers to the problem and they may be able to provide further assistance.
I hope this helps! If you have any other questions, feel free to ask.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
sand aspen
@reef lantern None of this worked
reef lanternBOT
Let me check the knowledge base for you. One moment! :books:
I'm sorry to hear that you're still having trouble with ESRGAN upscaling using the GPU in InvokeAI. Unfortunately, the information provided in the extracts does not contain a direct solution to this specific issue.
I recommend reaching out to the InvokeAI support team directly for further assistance. You can email them at support email (https://support.invoke.ai/cdn-cgi/l/email-protection#dd8ea8adadb2afa99db4b3abb2b6b8f3bcb4). They may be able to provide more specific guidance or troubleshooting steps.
Additionally, you could also report this issue on the InvokeAI GitHub page (https://github.com/invoke-ai/InvokeAI/issues). The developers or other users of the platform might have encountered similar issues and could provide further insights or solutions.