Issue After Installing AI Model

[plyght]

i got this message after it said initializing AI model: " Error: Unable to load dynamic library: Unable to load dynamic server library: dlopen(/var/folders/r3/41cmfh2s0l52ndz3w971w7lc0000gn/T/ollama2523156174/metal/libext_server.dylib, 0x0006): tried: '/var/folders/r3/41cmfh2s0l52ndz3w971w7lc0000gn/T/ollama2523156174/metal/libext_server.dylib' (code signature in <5A0BFDE3-DAFF-3CDC-88E1-EB69C3900B6B> '/private/var/folders/r3/41cmfh2s0l52ndz3w971w7lc0000gn/T/ollama2523156174/metal/libext_server.dylib' not valid for use in process: mapping process and mapped file (non-platform) have different Team IDs), "

[BruceMacD]

Hi @plyght, sorry about this. Are you running Ollama locally as well? I think this could be related to a bug in Ollama where it doesn't clean up its files from the temp directory.

[ag2307]

Hi, I wanted to try chatd and during the first run I also got the same error. I am not running any other LLM application locally. What could cause this issue?

Here's the service log (if required):

info: Ollama is not running: TypeError: fetch failed
info: Ollama is not installed on the system: TypeError [ERR_INVALID_ARG_TYPE]: The "path" argument must be of type string or an instance of Buffer or URL. Received undefined
info: waiting for ollama server...
info: fetch failed {"cause":{"address":"127.0.0.1","code":"ECONNREFUSED","errno":-61,"port":11434,"syscall":"connect"}}
info: ollama server is running
info: pulling model: mistral

[didikrm]

i got this message after it said initializing AI model: " Error: Unable to load dynamic library: Unable to load dynamic server library: dlopen(/var/folders/r3/41cmfh2s0l52ndz3w971w7lc0000gn/T/ollama2523156174/metal/libext_server.dylib, 0x0006): tried: '/var/folders/r3/41cmfh2s0l52ndz3w971w7lc0000gn/T/ollama2523156174/metal/libext_server.dylib' (code signature in <5A0BFDE3-DAFF-3CDC-88E1-EB69C3900B6B> '/private/var/folders/r3/41cmfh2s0l52ndz3w971w7lc0000gn/T/ollama2523156174/metal/libext_server.dylib' not valid for use in process: mapping process and mapped file (non-platform) have different Team IDs), "

I had the same issue. I reinstalled the app, went into MacOS terminal, typed ollama pull yourmodelname, then started chatd, and it worked.

[BenjaminStular]

Same issue (MacOS 14.1). Never installed Ollama, but have installed LMstudio and GPT4all (both probably/possibly using some of the Ollama's libraries). Both are closed, restarted, tried the Terminal - ollama pull yourmodelname, no change. Any ideas?

[BruceMacD]

I believe this should be fixed in the new v1.1.1 release.