Whisper not loaded and failed to load LLM

[macoron]

Whisper not loaded and failed to load LLM

[Macoron]

From this logs, it's hard to say what caused this error. Did you downloaded gpt4all-j-v1.3-groovy.bin model and put it StreamingAssets folder? Could you redownload it to make sure it's not corrupted?

Could you send log file last messages? Maybe there is some clues.

The whisper error is just my lazy copy-paste from another project.

[arcsidian]

Same error here on Unity 2022.3.2f1

I just started a fresh project, imported the package, put a working groovy model (tested) in StreamingAssets/Gpt4All and ran the project, I got the same errors, The first 2 errors show instantly, the 3rd error regarding whisper model shows up if I type something and submit it.

Error1:

Failed to load LLM model!
UnityEngine.Debug:LogError (object)
Gpt4All.LlmWrapper:InitFromPath (Gpt4All.LlmModelType,string) (at ./Packages/com.gpt4all.unity/Scripts/LlmWrapper.cs:202)
Gpt4All.LlmWrapper/<>c__DisplayClass28_0:b__0 () (at ./Packages/com.gpt4all.unity/Scripts/LlmWrapper.cs:212)
System.Threading._ThreadPoolWaitCallback:PerformWaitCallback ()

Error2:

Whisper model isn't loaded! Init Whisper model first!
UnityEngine.Debug:LogError (object)
Gpt4All.LlmManager/d__39:MoveNext () (at ./Packages/com.gpt4all.unity/Scripts/LlmManager.cs:142)
System.Runtime.CompilerServices.AsyncTaskMethodBuilder1<bool>:Start<Gpt4All.LlmManager/<CheckIfLoaded>d__39> (Gpt4All.LlmManager/<CheckIfLoaded>d__39&) Gpt4All.LlmManager:CheckIfLoaded () Gpt4All.LlmManager/<Prompt>d__38:MoveNext () (at ./Packages/com.gpt4all.unity/Scripts/LlmManager.cs:129) System.Runtime.CompilerServices.AsyncTaskMethodBuilder1:Start<Gpt4All.LlmManager/d__38> (Gpt4All.LlmManager/d__38&)
Gpt4All.LlmManager:Prompt (string)
Gpt4All.Samples.ChatSample/d__9:MoveNext () (at Assets/Samples/1 - Chat/ChatSample.cs:47)
System.Runtime.CompilerServices.AsyncVoidMethodBuilder:Start<Gpt4All.Samples.ChatSample/d__9> (Gpt4All.Samples.ChatSample/d__9&)
Gpt4All.Samples.ChatSample:SendToChat (string)
Gpt4All.Samples.ChatSample:OnSubmitPressed () (at Assets/Samples/1 - Chat/ChatSample.cs:35)
UnityEngine.EventSystems.EventSystem:Update () (at ./Library/PackageCache/[email protected]/Runtime/EventSystem/EventSystem.cs:530)

3rd error after pressing submit:

Whisper model isn't loaded! Init Whisper model first!
UnityEngine.Debug:LogError (object)
Gpt4All.LlmManager/d__39:MoveNext () (at ./Packages/com.gpt4all.unity/Scripts/LlmManager.cs:142)
System.Runtime.CompilerServices.AsyncTaskMethodBuilder1<bool>:Start<Gpt4All.LlmManager/<CheckIfLoaded>d__39> (Gpt4All.LlmManager/<CheckIfLoaded>d__39&) Gpt4All.LlmManager:CheckIfLoaded () Gpt4All.LlmManager/<Prompt>d__38:MoveNext () (at ./Packages/com.gpt4all.unity/Scripts/LlmManager.cs:129) System.Runtime.CompilerServices.AsyncTaskMethodBuilder1:Start<Gpt4All.LlmManager/d__38> (Gpt4All.LlmManager/d__38&)
Gpt4All.LlmManager:Prompt (string)
Gpt4All.Samples.ChatSample/d__9:MoveNext () (at Assets/Samples/1 - Chat/ChatSample.cs:47)
System.Runtime.CompilerServices.AsyncVoidMethodBuilder:Start<Gpt4All.Samples.ChatSample/d__9> (Gpt4All.Samples.ChatSample/d__9&)
Gpt4All.Samples.ChatSample:SendToChat (string)
Gpt4All.Samples.ChatSample:OnSubmitPressed () (at Assets/Samples/1 - Chat/ChatSample.cs:35)
UnityEngine.EventSystems.EventSystem:Update () (at ./Library/PackageCache/[email protected]/Runtime/EventSystem/EventSystem.cs:530)

[Macoron]

@arcsidian Does your editor log file contains any additional info? Can't reproduce it on my side with fresh install.

It should look like this or something similar:

Trying to load LLM model from C:/Users/dev/TestProject/Assets/StreamingAssets\Gpt4All/test.bin...
UnityEngine.StackTraceUtility:ExtractStackTrace ()
UnityEngine.DebugLogHandler:LogFormat (UnityEngine.LogType,UnityEngine.Object,string,object[])
UnityEngine.Logger:Log (UnityEngine.LogType,object)
UnityEngine.Debug:Log (object)
Gpt4All.LlmWrapper:InitFromPath (Gpt4All.LlmModelType,string) (at ./Library/PackageCache/com.gpt4all.unity@905af96848/Scripts/LlmWrapper.cs:194)
Gpt4All.LlmWrapper/<>c__DisplayClass28_0:<InitFromPathAsync>b__0 () (at ./Library/PackageCache/com.gpt4all.unity@905af96848/Scripts/LlmWrapper.cs:212)
System.Threading.Tasks.Task`1<Gpt4All.LlmWrapper>:InnerInvoke ()
System.Threading.Tasks.Task:Execute ()
System.Threading.Tasks.Task:ExecutionContextCallback (object)
System.Threading.ExecutionContext:RunInternal (System.Threading.ExecutionContext,System.Threading.ContextCallback,object,bool)
System.Threading.ExecutionContext:Run (System.Threading.ExecutionContext,System.Threading.ContextCallback,object,bool)
System.Threading.Tasks.Task:ExecuteWithThreadLocal (System.Threading.Tasks.Task&)
System.Threading.Tasks.Task:ExecuteEntry (bool)
System.Threading.Tasks.Task:System.Threading.IThreadPoolWorkItem.ExecuteWorkItem ()
System.Threading.ThreadPoolWorkQueue:Dispatch ()
System.Threading._ThreadPoolWaitCallback:PerformWaitCallback ()

(Filename: ./Library/PackageCache/com.gpt4all.unity@905af96848/Scripts/LlmWrapper.cs Line: 194)

gptj_model_load: loading model from 'C:/Users/dev/TestProject/Assets/StreamingAssets\Gpt4All/test.bin' - please wait ...
gptj_model_load: invalid model file 'C:/Users/dev/TestProject/Assets/StreamingAssets\Gpt4All/test.bin' (bad magic)
GPT-J ERROR: failed to load model from C:/Users/dev/TestProject/Assets/StreamingAssets\Gpt4All/test.binFailed to load LLM model!
UnityEngine.StackTraceUtility:ExtractStackTrace ()
UnityEngine.DebugLogHandler:LogFormat (UnityEngine.LogType,UnityEngine.Object,string,object[])
UnityEngine.Logger:Log (UnityEngine.LogType,object)
UnityEngine.Debug:LogError (object)
Gpt4All.LlmWrapper:InitFromPath (Gpt4All.LlmModelType,string) (at ./Library/PackageCache/com.gpt4all.unity@905af96848/Scripts/LlmWrapper.cs:202)
Gpt4All.LlmWrapper/<>c__DisplayClass28_0:<InitFromPathAsync>b__0 () (at ./Library/PackageCache/com.gpt4all.unity@905af96848/Scripts/LlmWrapper.cs:212)
System.Threading.Tasks.Task`1<Gpt4All.LlmWrapper>:InnerInvoke ()
System.Threading.Tasks.Task:Execute ()
System.Threading.Tasks.Task:ExecutionContextCallback (object)
System.Threading.ExecutionContext:RunInternal (System.Threading.ExecutionContext,System.Threading.ContextCallback,object,bool)
System.Threading.ExecutionContext:Run (System.Threading.ExecutionContext,System.Threading.ContextCallback,object,bool)
System.Threading.Tasks.Task:ExecuteWithThreadLocal (System.Threading.Tasks.Task&)
System.Threading.Tasks.Task:ExecuteEntry (bool)
System.Threading.Tasks.Task:System.Threading.IThreadPoolWorkItem.ExecuteWorkItem ()
System.Threading.ThreadPoolWorkQueue:Dispatch ()
System.Threading._ThreadPoolWaitCallback:PerformWaitCallback ()

[arcsidian]

Nevermind, it turned out to be a bad model, my bad. I redownloaded it and it works perfectly now!

[Macoron]

In original message @arcsidian had this error:

gptj_model_load: loading model from 'C:/Development/ProjectGPT/Assets/StreamingAssets\Gpt4All/ggml-gpt4all-j-v1.3-groovy.bin' - please wait ...
gptj_model_load: invalid model file 'C:/Development/ProjectGPT/Assets/StreamingAssets\Gpt4All/ggml-gpt4all-j-v1.3-groovy.bin' (bad magic)

Bad magic means that beginning of the file damaged or unknown. Hard to tell what might caused that, probably just download error.

@Yumiga check if redownloading fix that for you.

[Yumiga]

Okaaaaay ! I'll do it right know and tell ya !!!!!!!!!👌😁😁😁😁😊🤩

[Yumiga]

I redownloaded snoozy model and I got the same issues. Now, i'll check the groovy one.

Edit : Okay, it worked with groovy ! So I think it depend on the model we use...Thanks for help !🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤯🤯🤯🤯🤯🤯🤯🤪🤪🤪🤪🤪🤪🤪

[Macoron]

I redownloaded snoozy model and I got the same issues. Now, i'll check the groovy one.

Edit : Okay, it worked with groovy ! So I think it depend on the model we use...Thanks for help !🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤯🤯🤯🤯🤯🤯🤯🤪🤪🤪🤪🤪🤪🤪

FYI, groovy has Llama architecture. You should select LLAMA in dropdown Model Type in Llm Manager. Also don't use quantized versions, they are not supported yet.

[Yumiga]

I redownloaded snoozy model and I got the same issues. Now, i'll check the groovy one.
Edit : Okay, it worked with groovy ! So I think it depend on the model we use...Thanks for help !🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤩🤯🤯🤯🤯🤯🤯🤯🤪🤪🤪🤪🤪🤪🤪

FYI, groovy has Llama architecture. You should select LLAMA in dropdown Model Type in Llm Manager. Also don't use quantized versions, they are not supported yet.

Woops ! 😅 You 're right ! All issues are now solved !!!!! hehe !!!!!!! Thanks for ALL Macoron !!!!!!!!!!!!!!