Standardize model aliases and fix runtime bugs across all samples

samuel100 · Copilot · samuel100 · commit a84f7f7c7938 · 2026-03-30T15:10:52.000+01:00
- Use qwen2.5-0.5b for all chat samples (was phi-3.5-mini)
- Use whisper-tiny for all audio samples (was invalid 'whisper' alias)
- Use foundry_local_samples as common app_name across Python samples
- Fix tool-calling dict serialization bug in Python (ChatCompletionMessage -&gt; dict)
- Fix /v1 URL suffix for web-server and langchain-integration Python samples

Co-authored-by: Copilot &lt;223556219+Copilot@users.noreply.github.com&gt;
diff --git a/samples/cs/GettingStarted/src/TutorialChatAssistant/Program.cs b/samples/cs/GettingStarted/src/TutorialChatAssistant/Program.cs
@@ -26,7 +26,7 @@
 
 // Select and load a model from the catalog
 var catalog = await mgr.GetCatalogAsync();
-var model = await catalog.GetModelAsync("phi-3.5-mini")
+var model = await catalog.GetModelAsync("qwen2.5-0.5b")
     ?? throw new Exception("Model not found");
 
 await model.DownloadAsync(progress =>
diff --git a/samples/cs/GettingStarted/src/TutorialDocumentSummarizer/Program.cs b/samples/cs/GettingStarted/src/TutorialDocumentSummarizer/Program.cs
@@ -26,7 +26,7 @@
 
 // Select and load a model from the catalog
 var catalog = await mgr.GetCatalogAsync();
-var model = await catalog.GetModelAsync("phi-3.5-mini")
+var model = await catalog.GetModelAsync("qwen2.5-0.5b")
     ?? throw new Exception("Model not found");
 
 await model.DownloadAsync(progress =>
diff --git a/samples/cs/GettingStarted/src/TutorialToolCalling/Program.cs b/samples/cs/GettingStarted/src/TutorialToolCalling/Program.cs
@@ -123,7 +123,7 @@ string ExecuteTool(string functionName, JsonElement arguments)
 var mgr = FoundryLocalManager.Instance;
 
 var catalog = await mgr.GetCatalogAsync();
-var model = await catalog.GetModelAsync("phi-3.5-mini")
+var model = await catalog.GetModelAsync("qwen2.5-0.5b")
     ?? throw new Exception("Model not found");
 
 await model.DownloadAsync(progress =>
diff --git a/samples/cs/GettingStarted/src/TutorialVoiceToText/Program.cs b/samples/cs/GettingStarted/src/TutorialVoiceToText/Program.cs
@@ -31,7 +31,7 @@
 
 // <transcription>
 // Load the speech-to-text model
-var speechModel = await catalog.GetModelAsync("whisper")
+var speechModel = await catalog.GetModelAsync("whisper-tiny")
     ?? throw new Exception("Speech model not found");
 
 await speechModel.DownloadAsync(progress =>
@@ -63,7 +63,7 @@ await speechModel.DownloadAsync(progress =>
 
 // <summarization>
 // Load the chat model for summarization
-var chatModel = await catalog.GetModelAsync("phi-3.5-mini")
+var chatModel = await catalog.GetModelAsync("qwen2.5-0.5b")
     ?? throw new Exception("Chat model not found");
 
 await chatModel.DownloadAsync(progress =>
diff --git a/samples/js/tutorial-chat-assistant/app.js b/samples/js/tutorial-chat-assistant/app.js
@@ -12,7 +12,7 @@ const manager = FoundryLocalManager.create({
 });
 
 // Select and load a model from the catalog
-const model = await manager.catalog.getModel('phi-3.5-mini');
+const model = await manager.catalog.getModel('qwen2.5-0.5b');
 
 await model.download((progress) => {
     process.stdout.write(`\rDownloading model: ${progress.toFixed(2)}%`);
diff --git a/samples/js/tutorial-document-summarizer/app.js b/samples/js/tutorial-document-summarizer/app.js
@@ -41,7 +41,7 @@ const manager = FoundryLocalManager.create({
 });
 
 // Select and load a model from the catalog
-const model = await manager.catalog.getModel('phi-3.5-mini');
+const model = await manager.catalog.getModel('qwen2.5-0.5b');
 
 await model.download((progress) => {
     process.stdout.write(`\rDownloading model: ${progress.toFixed(2)}%`);
diff --git a/samples/js/tutorial-tool-calling/app.js b/samples/js/tutorial-tool-calling/app.js
@@ -120,7 +120,7 @@ const manager = FoundryLocalManager.create({
     logLevel: 'info'
 });
 
-const model = await manager.catalog.getModel('phi-3.5-mini');
+const model = await manager.catalog.getModel('qwen2.5-0.5b');
 
 await model.download((progress) => {
     process.stdout.write(
diff --git a/samples/js/tutorial-voice-to-text/app.js b/samples/js/tutorial-voice-to-text/app.js
@@ -13,7 +13,7 @@ const manager = FoundryLocalManager.create({
 
 // <transcription>
 // Load the speech-to-text model
-const speechModel = await manager.catalog.getModel('whisper');
+const speechModel = await manager.catalog.getModel('whisper-tiny');
 await speechModel.download((progress) => {
     process.stdout.write(
         `\rDownloading speech model: ${progress.toFixed(2)}%`
@@ -37,7 +37,7 @@ await speechModel.unload();
 
 // <summarization>
 // Load the chat model for summarization
-const chatModel = await manager.catalog.getModel('phi-3.5-mini');
+const chatModel = await manager.catalog.getModel('qwen2.5-0.5b');
 await chatModel.download((progress) => {
     process.stdout.write(
         `\rDownloading chat model: ${progress.toFixed(2)}%`
diff --git a/samples/python/audio-transcription/src/app.py b/samples/python/audio-transcription/src/app.py
@@ -12,7 +12,7 @@
 manager = FoundryLocalManager.instance
 
 # Load the whisper model for speech-to-text
-model = manager.catalog.get_model("whisper")
+model = manager.catalog.get_model("whisper-tiny")
 model.download(
     lambda progress: print(
         f"\rDownloading model: {progress:.2f}%",
diff --git a/samples/python/langchain-integration/src/app.py b/samples/python/langchain-integration/src/app.py
@@ -27,7 +27,7 @@
 
 # Start the web service to expose an OpenAI-compatible endpoint
 manager.start_web_service()
-base_url = manager.urls[0]
+base_url = f"{manager.urls[0]}/v1"
 # </init>
 
 # <langchain_setup>
diff --git a/samples/python/tool-calling/src/app.py b/samples/python/tool-calling/src/app.py
@@ -89,8 +89,23 @@ def process_tool_calls(messages, response, client):
     choice = response.choices[0].message
 
     while choice.tool_calls:
-        # Add the assistant message with tool call requests
-        messages.append(choice)
+        # Convert the assistant message to a dict for the SDK
+        assistant_msg = {
+            "role": "assistant",
+            "content": choice.content,
+            "tool_calls": [
+                {
+                    "id": tc.id,
+                    "type": tc.type,
+                    "function": {
+                        "name": tc.function.name,
+                        "arguments": tc.function.arguments,
+                    },
+                }
+                for tc in choice.tool_calls
+            ],
+        }
+        messages.append(assistant_msg)
 
         for tool_call in choice.tool_calls:
             function_name = tool_call.function.name
diff --git a/samples/python/tutorial-tool-calling/src/app.py b/samples/python/tutorial-tool-calling/src/app.py
@@ -89,8 +89,23 @@ def process_tool_calls(messages, response, client):
     choice = response.choices[0].message
 
     while choice.tool_calls:
-        # Add the assistant message with tool call requests
-        messages.append(choice)
+        # Convert the assistant message to a dict for the SDK
+        assistant_msg = {
+            "role": "assistant",
+            "content": choice.content,
+            "tool_calls": [
+                {
+                    "id": tc.id,
+                    "type": tc.type,
+                    "function": {
+                        "name": tc.function.name,
+                        "arguments": tc.function.arguments,
+                    },
+                }
+                for tc in choice.tool_calls
+            ],
+        }
+        messages.append(assistant_msg)
 
         for tool_call in choice.tool_calls:
             function_name = tool_call.function.name
diff --git a/samples/python/tutorial-voice-to-text/src/app.py b/samples/python/tutorial-voice-to-text/src/app.py
@@ -15,7 +15,7 @@ async def main():
 
     # <transcription>
     # Load the speech-to-text model
-    speech_model = manager.catalog.get_model("whisper")
+    speech_model = manager.catalog.get_model("whisper-tiny")
     speech_model.download(
         lambda progress: print(
             f"\rDownloading speech model: {progress:.2f}%",
diff --git a/samples/python/web-server/src/app.py b/samples/python/web-server/src/app.py
@@ -25,7 +25,7 @@
 
 # Start the web service to expose an OpenAI-compatible REST endpoint
 manager.start_web_service()
-base_url = manager.urls[0]
+base_url = f"{manager.urls[0]}/v1"
 # </init>
 
 # <rest_client>
diff --git a/samples/rust/tutorial-chat-assistant/src/main.rs b/samples/rust/tutorial-chat-assistant/src/main.rs
@@ -16,7 +16,7 @@ async fn main() -> anyhow::Result<()> {
     let manager = FoundryLocalManager::create(FoundryLocalConfig::new("chat-assistant"))?;
 
     // Select and load a model from the catalog
-    let model = manager.catalog().get_model("phi-3.5-mini").await?;
+    let model = manager.catalog().get_model("qwen2.5-0.5b").await?;
 
     if !model.is_cached().await? {
         println!("Downloading model...");
diff --git a/samples/rust/tutorial-document-summarizer/src/main.rs b/samples/rust/tutorial-document-summarizer/src/main.rs
@@ -90,7 +90,7 @@ async fn main() -> anyhow::Result<()> {
     // Select and load a model from the catalog
     let model = manager
         .catalog()
-        .get_model("phi-3.5-mini")
+        .get_model("qwen2.5-0.5b")
         .await?;
 
     if !model.is_cached().await? {
diff --git a/samples/rust/tutorial-tool-calling/src/main.rs b/samples/rust/tutorial-tool-calling/src/main.rs
@@ -193,7 +193,7 @@ async fn main() -> anyhow::Result<()> {
     // Select and load a model
     let model = manager
         .catalog()
-        .get_model("phi-3.5-mini")
+        .get_model("qwen2.5-0.5b")
         .await?;
 
     if !model.is_cached().await? {
diff --git a/samples/rust/tutorial-voice-to-text/src/main.rs b/samples/rust/tutorial-voice-to-text/src/main.rs
@@ -22,7 +22,7 @@ async fn main() -> anyhow::Result<()> {
     // Load the speech-to-text model
     let speech_model = manager
         .catalog()
-        .get_model("whisper")
+        .get_model("whisper-tiny")
         .await?;
 
     if !speech_model.is_cached().await? {
@@ -54,7 +54,7 @@ async fn main() -> anyhow::Result<()> {
     // Load the chat model for summarization
     let chat_model = manager
         .catalog()
-        .get_model("phi-3.5-mini")
+        .get_model("qwen2.5-0.5b")
         .await?;
 
     if !chat_model.is_cached().await? {
