Phi-4-mini-reasoning crash on Qualcomm device

[kinfey]

if I run foundry local in Qualcomm using Phi-4-mini-reasoning , it will crash.

My vm has 64GB memory(https://qdc.qualcomm.com/) , but Deepseek all models are okay .

My devices

My NPU

The log is

2025-06-03 01:22:59.509 -07:00 [INF] Starting Foundry Local CLI with 'model run Phi-4-mini-reasoning-qnn-npu'
2025-06-03 01:22:59.516 -07:00 [INF] Starting service <C:\Program Files\WindowsApps\Microsoft.FoundryLocal_0.3.9267.43123_arm64__8wekyb3d8bbwe\Inference.Service.Agent.exe --urls=http://localhost:5273 --OpenAIServiceSettings:ModelDirPath="C:\Users\HCKTest.foundry\cache\models" --JsonRpcServer:Run=true --JsonRpcServer:PipeName="inference_agent">
2025-06-03 01:22:59.625 -07:00 [INF] Command:ServiceStart Status:Success Direct:False Time:111ms
2025-06-03 01:23:00.366 -07:00 [INF] Downloading Phi-4-mini-reasoning-qnn-npu url:http://localhost:5273/openai/download
2025-06-03 01:24:04.359 -07:00 [INF] Command:ModelDownload Status:Success Direct:False Time:64733ms
2025-06-03 01:24:04.359 -07:00 [INF] Loading model: http://localhost:5273/openai/load/Phi-4-mini-reasoning-qnn-npu?ttl=600
2025-06-03 01:24:09.849 -07:00 [INF] 🟢 Model Phi-4-mini-reasoning-qnn-npu loaded successfully
2025-06-03 01:24:09.849 -07:00 [INF] Command:ModelLoad Status:Success Direct:False Time:5489ms
2025-06-03 01:24:47.343 -07:00 [INF] Command:SessionInteractive Status:Failure Direct:False Time:37492ms
2025-06-03 01:24:47.343 -07:00 [INF] Command:ModelRun Status:Failure Direct:True Time:107827ms
2025-06-03 01:24:47.346 -07:00 [INF] LogException
Microsoft.AI.Foundry.Local.Common.FLException: Error during chat
---> System.Net.Http.HttpIOException: The response ended prematurely. (ResponseEnded)
at System.Net.Http.HttpConnection.d__82.MoveNext() + 0x288
--- End of stack trace from previous location ---
at System.Net.Http.HttpConnection.ChunkedEncodingReadStream.d__6.MoveNext() + 0x288
--- End of stack trace from previous location ---
at System.ClientModel.Internal.ReadTimeoutStream.d__21.MoveNext() + 0x298
--- End of stack trace from previous location ---
at System.Net.ServerSentEvents.SseParser1.<FillLineBufferAsync>d__35.MoveNext() + 0x1d8 --- End of stack trace from previous location --- at System.Net.ServerSentEvents.SseParser1.d__20.MoveNext() + 0x4fc
--- End of stack trace from previous location ---
at OpenAI.AsyncSseUpdateCollection1.AsyncSseUpdateEnumerator1.<System-Collections-Generic-IAsyncEnumerator-MoveNextAsync>d__12.MoveNext() + 0x1f4
--- End of stack trace from previous location ---
at OpenAI.AsyncSseUpdateCollection1.<GetValuesFromPageAsync>d__8.MoveNext() + 0x270 --- End of stack trace from previous location --- at OpenAI.AsyncSseUpdateCollection1.d__8.MoveNext() + 0x3e8
--- End of stack trace from previous location ---
at System.ClientModel.AsyncCollectionResult1.<GetAsyncEnumerator>d__1.MoveNext() + 0x300 --- End of stack trace from previous location --- at System.ClientModel.AsyncCollectionResult1.d__1.MoveNext() + 0x6bc
--- End of stack trace from previous location ---
at System.ClientModel.AsyncCollectionResult1.<GetAsyncEnumerator>d__1.MoveNext() + 0x800 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<InteractiveNewRoundAsync>d__4.MoveNext() + 0x1e4 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<InteractiveNewRoundAsync>d__4.MoveNext() + 0x4bc --- End of inner exception stack trace --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.HandleExceptionDuringChat(Exception) + 0x108 at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<InteractiveNewRoundAsync>d__4.MoveNext() + 0x65c --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<RunInteractiveNewRoundAsync>d__2.MoveNext() + 0x64 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<<Create>b__1_0>d.MoveNext() + 0x1694 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Common.CommandActionFactory.<>c__DisplayClass0_01.<b__0>d.MoveNext() + 0x238
--- End of stack trace from previous location ---
at System.CommandLine.NamingConventionBinder.CommandHandler.d__66.MoveNext() + 0x5c
--- End of stack trace from previous location ---
at System.CommandLine.NamingConventionBinder.ModelBindingCommandHandler.d__11.MoveNext() + 0x6c
--- End of stack trace from previous location ---
at System.CommandLine.Invocation.InvocationPipeline.d__0.MoveNext() + 0x1f4
--- End of stack trace from previous location ---
at Microsoft.AI.Foundry.Local.Program.

d__1.MoveNext() + 0x52c

[leestott]

Confirming this issue is using the Qualcomm Devbox https://qdc.qualcomm.com/ so this is VM developer machine provided by Qualcomm

Build Details
Snapdragon® X Elite | CRD8380X
Hamoa.WP.1.0.c90-07350-STD.PROD-1

Here is my test

using the cpu model this works ok its a issue with NPU Model build

Log

Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:12:38.98 0 OpenAIService for type:OpenAIServiceProviderOnnx
Information: Microsoft.Neutron.Rpc.Service.JsonRpcService[0]
02:12:38.98 0 json-rpc-service running
Information: Microsoft.Neutron.Rpc.Service.JsonRpcService[2305]
02:12:38.99 accept_pipe_connections Accepting pipe connections pipeName:inference_agent
Information: Microsoft.Hosting.Lifetime[14]
02:12:38.99 ListeningOnAddress Now listening on: http://localhost:5273
Information: Microsoft.Hosting.Lifetime[0]
02:12:38.99 0 Application started. Press Ctrl+C to shut down.
Information: Microsoft.Hosting.Lifetime[0]
02:12:38.99 0 Hosting environment: Production
Information: Microsoft.Hosting.Lifetime[0]
02:12:38.99 0 Content root path: C:\Program Files\WindowsApps\Microsoft.FoundryLocal_0.4.91.9885_arm64__8wekyb3d8bbwe
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.05 0 Valid devices: cpu, npu
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.05 0 Valid EPs: cpuexecutionprovider, qnnexecutionprovider
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.78 0 Model Phi-4-reasoning-generic-cpu does not have a valid prompt template.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.78 0 Model deepseek-r1-distill-qwen-14b-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.78 0 Model deepseek-r1-distill-qwen-7b-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.78 0 Model Phi-4-mini-instruct-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:12:39.78 0 Total models fetched across all pages: 17
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:13:44.12 0 Model Phi-4-reasoning-generic-cpu does not have a valid prompt template.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:13:44.12 0 Model deepseek-r1-distill-qwen-14b-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:13:44.12 0 Model deepseek-r1-distill-qwen-7b-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:13:44.12 0 Model Phi-4-mini-instruct-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:13:44.12 0 Total models fetched across all pages: 17
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:44.14 0 Azure Download model Uri:azureml://registries/azureml/models/Phi-4-mini-reasoning-qnn-npu/versions/1 path:(null) revision:(null) outputDirectory:C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/LICENSE|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/LICENSE
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/README.md|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/README.md
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/added_tokens.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/added_tokens.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/config.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/genai_config.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/genai_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/merges.txt|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/merges.txt
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_1.bin|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_1.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_2.bin|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_2.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_3.bin|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_3.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_4.bin|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_4.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_ctx.onnx_ctx.onnx|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_ctx.onnx_ctx.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_embeddings.all.quant.onnx|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_embeddings.all.quant.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_iter.onnx_ctx.onnx|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_iter.onnx_ctx.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/phi_4_mini_lm_head.all.quant.onnx|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/phi_4_mini_lm_head.all.quant.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/special_tokens_map.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/special_tokens_map.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/tokenizer.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/tokenizer.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/tokenizer_config.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/tokenizer_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 qnn-phi4-mini-reasoning-onnx/vocab.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-qnn-npu\qnn-phi4-mini-reasoning-onnx/vocab.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 Azure Download totalSizeToDownload:3004831133
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/LICENSE contentLength:1105
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/README.md contentLength:18370
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/added_tokens.json contentLength:249
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.34 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/config.json contentLength:2504
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/genai_config.json contentLength:17026
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/merges.txt contentLength:2418348
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_1.bin contentLength:508554120
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_2.bin contentLength:422030096
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_3.bin contentLength:409443200
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_4.bin contentLength:409451600
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_ctx.onnx_ctx.onnx contentLength:1630519
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_embeddings.all.quant.onnx contentLength:614597133
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_iter.onnx_ctx.onnx contentLength:1630492
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_lm_head.all.quant.onnx contentLength:615597957
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/special_tokens_map.json contentLength:587
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/tokenizer.json contentLength:15524194
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/tokenizer_config.json contentLength:3323
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.37 0 Azure Downloading blob name:qnn-phi4-mini-reasoning-onnx/vocab.json contentLength:3910310
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.44 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/LICENSE
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.44 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/LICENSE
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.54 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/README.md
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.54 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/README.md
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.58 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/added_tokens.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.58 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/added_tokens.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.60 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/genai_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.62 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.62 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:46.62 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/genai_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:47.62 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/tokenizer_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:13:48.44 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/special_tokens_map.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:14:26.01 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/vocab.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:14:26.32 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/merges.txt
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:14:26.32 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/merges.txt
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:14:29.62 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_iter.onnx_ctx.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:14:48.87 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_ctx.onnx_ctx.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:00.68 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_3.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:01.35 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/tokenizer.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:03.83 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_lm_head.all.quant.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:04.33 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_4.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:05.68 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_2.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:10.38 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_1.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:10.38 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_1.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:10.38 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_2.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:10.38 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_3.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:10.38 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_cb_4.bin
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:10.38 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_ctx.onnx_ctx.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_embeddings.all.quant.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_embeddings.all.quant.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_iter.onnx_ctx.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/phi_4_mini_lm_head.all.quant.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/special_tokens_map.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/tokenizer.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/tokenizer_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:15:29.75 0 Azure Download Completed name:qnn-phi4-mini-reasoning-onnx/vocab.json
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[1400]
02:15:29.76 load_model_started Loading model:Phi-4-mini-reasoning-qnn-npu
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[1401]
02:15:35.56 load_model_finished Finish loading model:Phi-4-mini-reasoning-qnn-npu elapsed time:00:00:05.8011926
Information: Microsoft.Neutron.OpenAI.Delegates.OpenAIApi[0]
02:15:41.45 0 HandleChatCompletionAsStreamRequest -> model:Phi-4-mini-reasoning-qnn-npu MaxCompletionTokens:2048 maxTokens:(null) temperature:(null) topP:(null)
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:15:41.45 0 HandleChatCompletionAsStreamRequest -> model:Phi-4-mini-reasoning-qnn-npu MaxCompletionTokens:2048 maxTokens:(null) temperature:(null) topP:(null)
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:15:41.83 0 TTFT: 1 ms, TTST: 0
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:15:41.83 0 Completion stats: Total Tokens/Second: 2.6595744680851063, Time to First Token: 314 ms, Average Token Generation Time: 314 ms, Total Time: 376 ms, Total Tokens: 1
Error: Microsoft.AspNetCore.Server.Kestrel[13]
02:15:41.87 ApplicationError Connection id "0HND4R94FGQ9B", Request id "0HND4R94FGQ9B:00000001": An unhandled exception was thrown by the application. message:Non-zero status code returned while running QNN_1828573908117222568_55 node. Name:'QNNExecutionProvider_QNN_1828573908117222568_55_19' Status Message: QNN graph execute error. Error code: 6031 stack: at Microsoft.ML.OnnxRuntimeGenAI.Result.VerifySuccess(IntPtr) + 0x58
at Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderBase1.<>c__DisplayClass36_0.<<HandleChatCompletionAsStreamRequestAsync>g__DecodeTokens|0>d.MoveNext() + 0x130 --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.TokenPerfExtensions.<WithTokenPerfAsync>d__02.MoveNext() + 0x304
--- End of stack trace from previous location ---
at Microsoft.Neutron.OpenAI.TokenPerfExtensions.d__02.MoveNext() + 0x6d8 --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.TokenPerfExtensions.<WithTokenPerfAsync>d__02.MoveNext() + 0x848
--- End of stack trace from previous location ---
at Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderBase1.<HandleChatCompletionAsStreamRequestAsync>d__36.MoveNext() + 0x68c --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderBase1.d__36.MoveNext() + 0x838
--- End of stack trace from previous location ---
at Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderBase1.<HandleChatCompletionAsStreamRequestAsync>d__36.MoveNext() + 0x97c --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.OpenAIJsonExtensions.<WriteChatCompletionResponses>d__19.MoveNext() + 0x408 --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.OpenAIJsonExtensions.<WriteChatCompletionResponses>d__19.MoveNext() + 0x6d0 --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.OpenAIJsonExtensions.<WriteChatCompletionResponses>d__19.MoveNext() + 0x7e8 --- End of stack trace from previous location --- at Microsoft.Neutron.OpenAI.OpenAIServiceWebApiExtensions.<>c__DisplayClass0_0.<<HandleStreamRequest>b__0>d.MoveNext() + 0x5c --- End of stack trace from previous location --- at Microsoft.AspNetCore.Http.Generated.<GeneratedRouteBuilderExtensions_g>F16C589DE9EC82483AA705851D2FE201CB4CB4AAF6561E8DE71B6A1891AD8D67F__GeneratedRouteBuilderExtensionsCore.<>c__DisplayClass11_0.<<MapPost7>g__RequestHandler|5>d.MoveNext() + 0x20c --- End of stack trace from previous location --- at Microsoft.AspNetCore.Server.Kestrel.Core.Internal.Http.HttpProtocol.<ProcessRequests>d__2381.MoveNext() + 0x290
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:16:16.84 0 Model Phi-4-reasoning-generic-cpu does not have a valid prompt template.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:16:16.84 0 Model deepseek-r1-distill-qwen-14b-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:16:16.84 0 Model deepseek-r1-distill-qwen-7b-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:16:16.84 0 Model Phi-4-mini-instruct-generic-cpu is not supported on Arm64 currently.
Information: Microsoft.Neutron.AzureFoundry.AzureFoundryService[0]
02:16:16.84 0 Total models fetched across all pages: 17
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:16.85 0 Azure Download model Uri:azureml://registries/azureml/models/Phi-4-mini-reasoning-generic-cpu/versions/2 path:(null) revision:(null) outputDirectory:C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/added_tokens.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/added_tokens.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/genai_config.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/genai_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/merges.txt|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/merges.txt
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/model.onnx|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/model.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/model.onnx.data|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/model.onnx.data
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/special_tokens_map.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/special_tokens_map.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/tokenizer.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/tokenizer.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/tokenizer_config.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/tokenizer_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 v1/vocab.json|C:\Users\HCKTest.foundry\cache\models\Microsoft\Phi-4-mini-reasoning-generic-cpu\v1/vocab.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 Azure Download totalSizeToDownload:4930551501
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 Azure Downloading blob name:v1/added_tokens.json contentLength:249
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 Azure Downloading blob name:v1/genai_config.json contentLength:1520
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 Azure Downloading blob name:v1/merges.txt contentLength:2418348
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.75 0 Azure Downloading blob name:v1/model.onnx contentLength:52119126
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.77 0 Azure Downloading blob name:v1/model.onnx.data contentLength:4856573952
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.77 0 Azure Downloading blob name:v1/special_tokens_map.json contentLength:587
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.77 0 Azure Downloading blob name:v1/tokenizer.json contentLength:15524193
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.77 0 Azure Downloading blob name:v1/tokenizer_config.json contentLength:3216
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.77 0 Azure Downloading blob name:v1/vocab.json contentLength:3910310
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.83 0 Azure Download Completed name:v1/added_tokens.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.83 0 Azure Download Completed name:v1/added_tokens.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.91 0 Azure Download Completed name:v1/genai_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:17.91 0 Azure Download Completed name:v1/genai_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:18.03 0 Azure Download Completed name:v1/special_tokens_map.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:18.06 0 Azure Download Completed name:v1/tokenizer_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:18.99 0 Azure Download Completed name:v1/merges.txt
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:18.99 0 Azure Download Completed name:v1/merges.txt
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:31.38 0 Azure Download Completed name:v1/vocab.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:33.18 0 Azure Download Completed name:v1/model.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:33.18 0 Azure Download Completed name:v1/model.onnx
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:16:33.74 0 Azure Download Completed name:v1/tokenizer.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:18:16.22 0 Azure Download Completed name:v1/model.onnx.data
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:18:16.22 0 Azure Download Completed name:v1/model.onnx.data
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:18:16.22 0 Azure Download Completed name:v1/special_tokens_map.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:18:16.22 0 Azure Download Completed name:v1/tokenizer.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:18:16.22 0 Azure Download Completed name:v1/tokenizer_config.json
Information: Microsoft.Neutron.OpenAI.Provider.AzureBlobDownloadClientProvider[0]
02:18:16.22 0 Azure Download Completed name:v1/vocab.json
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[1403]
02:18:16.23 unload_model Unloading model:Phi-4-mini-reasoning-qnn-npu
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[1400]
02:18:17.40 load_model_started Loading model:Phi-4-mini-reasoning-generic-cpu
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[1401]
02:18:22.24 load_model_finished Finish loading model:Phi-4-mini-reasoning-generic-cpu elapsed time:00:00:04.8467633
Information: Microsoft.Neutron.OpenAI.Delegates.OpenAIApi[0]
02:18:29.65 0 HandleChatCompletionAsStreamRequest -> model:Phi-4-mini-reasoning-generic-cpu MaxCompletionTokens:2048 maxTokens:(null) temperature:(null) topP:(null)
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:18:29.65 0 HandleChatCompletionAsStreamRequest -> model:Phi-4-mini-reasoning-generic-cpu MaxCompletionTokens:2048 maxTokens:(null) temperature:(null) topP:(null)
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:19:27.99 0 TTFT: 1 ms, TTST: 31
Information: Microsoft.Neutron.OpenAI.Provider.OpenAIServiceProviderOnnx[0]
02:19:27.99 0 Completion stats: Total Tokens/Second: 24.478461353857757, Time to First Token: 170 ms, Average Token Generation Time: 40.25070028011204 ms, Total Time: 58337 ms, Total Tokens: 1428

and

2025-06-06 02:12:38.816 -07:00 [INF] Starting Foundry Local CLI with 'model list'
2025-06-06 02:12:38.839 -07:00 [INF] Starting service <C:\Program Files\WindowsApps\Microsoft.FoundryLocal_0.4.91.9885_arm64__8wekyb3d8bbwe\Inference.Service.Agent.exe --urls=http://localhost:5273 --OpenAIServiceSettings:ModelDirPath="C:\Users\HCKTest.foundry\cache\models" --JsonRpcServer:Run=true --JsonRpcServer:PipeName="inference_agent">
2025-06-06 02:12:39.799 -07:00 [INF] Command:ModelList Status:Success Direct:True Time:964ms
2025-06-06 02:13:43.525 -07:00 [INF] Starting Foundry Local CLI with 'model run phi-4-mini-reasoning-qnn-npu'
2025-06-06 02:13:43.531 -07:00 [INF] Command:ServiceStart Status:Skipped Direct:False Time:1ms
2025-06-06 02:13:44.140 -07:00 [INF] Downloading Phi-4-mini-reasoning-qnn-npu url:http://localhost:5273/openai/download
2025-06-06 02:15:29.755 -07:00 [INF] Command:ModelDownload Status:Success Direct:False Time:106229ms
2025-06-06 02:15:29.757 -07:00 [INF] Command:ServiceList Status:Success Direct:False Time:1ms
2025-06-06 02:15:29.757 -07:00 [INF] Loading model: http://localhost:5273/openai/load/Phi-4-mini-reasoning-qnn-npu?ttl=600
2025-06-06 02:15:35.565 -07:00 [INF] 🟢 Model Phi-4-mini-reasoning-qnn-npu loaded successfully
2025-06-06 02:15:35.565 -07:00 [INF] Command:ModelLoad Status:Success Direct:False Time:5808ms
2025-06-06 02:15:41.871 -07:00 [INF] Command:SessionInteractive Status:Failure Direct:False Time:6305ms
2025-06-06 02:15:41.871 -07:00 [INF] Command:ModelRun Status:Failure Direct:True Time:118348ms
2025-06-06 02:15:41.882 -07:00 [INF] LogException
Microsoft.AI.Foundry.Local.Common.FLException: Error during chat
---> System.Net.Http.HttpIOException: The response ended prematurely. (ResponseEnded)
at System.Net.Http.HttpConnection.d__82.MoveNext() + 0x288
--- End of stack trace from previous location ---
at System.Net.Http.HttpConnection.ChunkedEncodingReadStream.d__6.MoveNext() + 0x288
--- End of stack trace from previous location ---
at System.ClientModel.Internal.ReadTimeoutStream.d__21.MoveNext() + 0x298
--- End of stack trace from previous location ---
at System.Net.ServerSentEvents.SseParser1.<FillLineBufferAsync>d__35.MoveNext() + 0x1d8 --- End of stack trace from previous location --- at System.Net.ServerSentEvents.SseParser1.d__20.MoveNext() + 0x4fc
--- End of stack trace from previous location ---
at OpenAI.AsyncSseUpdateCollection1.AsyncSseUpdateEnumerator1.<System-Collections-Generic-IAsyncEnumerator-MoveNextAsync>d__12.MoveNext() + 0x1f4
--- End of stack trace from previous location ---
at OpenAI.AsyncSseUpdateCollection1.<GetValuesFromPageAsync>d__8.MoveNext() + 0x270 --- End of stack trace from previous location --- at OpenAI.AsyncSseUpdateCollection1.d__8.MoveNext() + 0x3e8
--- End of stack trace from previous location ---
at System.ClientModel.AsyncCollectionResult1.<GetAsyncEnumerator>d__1.MoveNext() + 0x300 --- End of stack trace from previous location --- at System.ClientModel.AsyncCollectionResult1.d__1.MoveNext() + 0x6bc
--- End of stack trace from previous location ---
at System.ClientModel.AsyncCollectionResult1.<GetAsyncEnumerator>d__1.MoveNext() + 0x800 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<InteractiveNewRoundAsync>d__5.MoveNext() + 0x1e0 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<InteractiveNewRoundAsync>d__5.MoveNext() + 0x4e0 --- End of inner exception stack trace --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.HandleExceptionDuringChat(Exception) + 0x108 at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<InteractiveNewRoundAsync>d__5.MoveNext() + 0x680 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<RunInteractiveNewRoundAsync>d__3.MoveNext() + 0x64 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Commands.ModelRunCommand.<<Create>b__1_0>d.MoveNext() + 0x1504 --- End of stack trace from previous location --- at Microsoft.AI.Foundry.Local.Common.CommandActionFactory.<>c__DisplayClass0_01.<b__0>d.MoveNext() + 0x238
--- End of stack trace from previous location ---
at System.CommandLine.NamingConventionBinder.CommandHandler.d__66.MoveNext() + 0x5c
--- End of stack trace from previous location ---
at System.CommandLine.NamingConventionBinder.ModelBindingCommandHandler.d__11.MoveNext() + 0x6c
--- End of stack trace from previous location ---
at System.CommandLine.Invocation.InvocationPipeline.d__0.MoveNext() + 0x1f4
--- End of stack trace from previous location ---
at Microsoft.AI.Foundry.Local.Program.

d__1.MoveNext() + 0x52c
2025-06-06 02:16:16.158 -07:00 [INF] Starting Foundry Local CLI with 'model run Phi-4-mini-reasoning-generic-cpu'
2025-06-06 02:16:16.165 -07:00 [INF] Command:ServiceStart Status:Skipped Direct:False Time:2ms
2025-06-06 02:16:16.856 -07:00 [INF] Downloading Phi-4-mini-reasoning-generic-cpu url:http://localhost:5273/openai/download
2025-06-06 02:18:16.234 -07:00 [INF] Command:ModelDownload Status:Success Direct:False Time:120075ms
2025-06-06 02:18:16.236 -07:00 [INF] Command:ServiceList Status:Success Direct:False Time:1ms
2025-06-06 02:18:17.393 -07:00 [INF] Command:ModelUnloadAll Status:Success Direct:False Time:1156ms
2025-06-06 02:18:17.393 -07:00 [INF] Loading model: http://localhost:5273/openai/load/Phi-4-mini-reasoning-generic-cpu?ttl=600
2025-06-06 02:18:22.250 -07:00 [INF] 🟢 Model Phi-4-mini-reasoning-generic-cpu loaded successfully
2025-06-06 02:18:22.250 -07:00 [INF] Command:ModelLoad Status:Success Direct:False Time:4857ms
2025-06-06 02:20:44.126 -07:00 [INF] Command:SessionInteractive Status:Success Direct:False Time:141883ms
2025-06-06 02:20:44.126 -07:00 [INF] Command:ModelRun Status:Success Direct:True Time:267978ms
2025-06-06 02:23:12.181 -07:00 [INF] Starting Foundry Local CLI with 'service diag --logs'

[natke]

@natke to try to repro