Substage does not run properly with local LLMs on my Mac. I used the settings to add Qween via Ollama. This model works well via the Ollama interface. However when I ask substage to do something, I get the following error message: "could not connect to the server"
I also am getting timeouts for most of my local model requests, even when the request is successful.
joethephish
Team•
May 21
Every case I’ve seen this going wrong has been relating to the model settings not being set up correctly: please double check the base URL - for Ollama that’s usually
Log in to comment and vote
Comments5
lskinransup
Jun 1
I had a similar problem fixed by using exactly the same name for the model as Ollama does.
jmfontana
May 23
OK, after I changed the localhost things seemed to work but now I’m constantly getting error messages saying “The request timed out”.
Here’s a sample of the log file:
[2026-05-23T11:59:51Z] INFO: PROMPT: Find duplicates in any of the subfolders., MODEL: Gemma 4 [2026-05-23T12:00:51Z] ERROR: AI request error: Error Domain=NSURLErrorDomain Code=-1001 "The request timed out." UserInfo={_kCFStreamErrorCodeKey=-2102, NSUnderlyingError=0xb792ce820 {Error Domain=kCFErrorDomainCFNetwork Code=-1001 "(null)" UserInfo={_kCFStreamErrorCodeKey=-2102, _kCFStreamErrorDomainKey=4}}, _NSURLErrorFailingURLSessionTaskErrorKey=LocalDataTask <B42C3D12-E9EB-47EE-A221-B64071871746>.<19>, _NSURLErrorRelatedURLSessionTaskErrorKey=( "LocalDataTask <B42C3D12-E9EB-47EE-A221-B64071871746>.<19>" ), NSLocalizedDescription=The request timed out., NSErrorFailingURLStringKey=http://localhost:11434/v1/chat/completions, NSErrorFailingURLKey=http://localhost:11434/v1/chat/completions, _kCFStreamErrorDomainKey=4} [2026-05-23T12:00:51Z] ERROR: ERROR: Error Domain=NSURLErrorDomain Code=-1001 "The request timed out." UserInfo={_kCFStreamErrorCodeKey=-2102, NSUnderlyingError=0xb792ce820 {Error Domain=kCFErrorDomainCFNetwork Code=-1001 "(null)" UserInfo={_kCFStreamErrorCodeKey=-2102, _kCFStreamErrorDomainKey=4}}, _NSURLErrorFailingURLSessionTaskErrorKey=LocalDataTask <B42C3D12-E9EB-47EE-A221-B64071871746>.<19>, _NSURLErrorRelatedURLSessionTaskErrorKey=( "LocalDataTask <B42C3D12-E9EB-47EE-A221-B64071871746>.<19>" ), NSLocalizedDescription=The request timed out., NSErrorFailingURLStringKey=http://localhost:11434/v1/chat/completions, NSErrorFailingURLKey=http://localhost:11434/v1/chat/completions, _kCFStreamErrorDomainKey=4}
lskinransup
Jun 1
I also am getting timeouts for most of my local model requests, even when the request is successful.
joethephish
May 21
Every case I’ve seen this going wrong has been relating to the model settings not being set up correctly: please double check the base URL - for Ollama that’s usually
http://localhost:11434 if you’re running it on the same computer.
jmfontana
May 21
Perfect. Yes. Changing the local host to http://localhost:11434 did the trick. Thanks.