Merge branch 'main' into main

Author: HanCreation

articles/openai-harmony.md

@@ -6,7 +6,7 @@ The [`gpt-oss` models](https://openai.com/open-models) were trained on the harmo
 
 ### Roles
 
-Every message that the model processes has a role associated with it. The model knows about three types of roles:
+Every message that the model processes has a role associated with it. The model knows about five types of roles:
 
 | Role        | Purpose                                                                                                                                                                                 |
 | :---------- | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
@@ -91,14 +91,14 @@ convo = Conversation.from_messages(
         Message.from_role_and_content(Role.USER, "What is the weather in Tokyo?"),
         Message.from_role_and_content(
             Role.ASSISTANT,
-            'User asks: "What is the weather in Tokyo?" We need to use get_weather tool.',
+            'User asks: "What is the weather in Tokyo?" We need to use get_current_weather tool.',
         ).with_channel("analysis"),
         Message.from_role_and_content(Role.ASSISTANT, '{"location": "Tokyo"}')
         .with_channel("commentary")
-        .with_recipient("functions.get_weather")
+        .with_recipient("functions.get_current_weather")
         .with_content_type("<|constrain|> json"),
         Message.from_author_and_content(
-            Author.new(Role.TOOL, "functions.lookup_weather"),
+            Author.new(Role.TOOL, "functions.get_current_weather"),
             '{ "temperature": 20, "sunny": true }',
         ).with_channel("commentary"),
     ]
@@ -376,7 +376,7 @@ If the model decides to call a tool it will define a `recipient` in the header o
 The model might also specify a `<|constrain|>` token to indicate the type of input for the tool call. In this case since it’s being passed in as JSON the `<|constrain|>` is set to `json`.
 
 ```
-<|channel|>analysis<|message|>Need to use function get_weather.<|end|><|start|>assistant<|channel|>commentary to=functions.get_weather <|constrain|>json<|message|>{"location":"San Francisco"}<|call|>
+<|channel|>analysis<|message|>Need to use function get_current_weather.<|end|><|start|>assistant<|channel|>commentary to=functions.get_current_weather <|constrain|>json<|message|>{"location":"San Francisco"}<|call|>
 ```
 
 #### Handling tool calls
@@ -392,7 +392,7 @@ A tool message has the following format:
 So in our example above
 
 ```
-<|start|>functions.get_weather to=assistant<|channel|>commentary<|message|>{"sunny": true, "temperature": 20}<|end|>
+<|start|>functions.get_current_weather to=assistant<|channel|>commentary<|message|>{"sunny": true, "temperature": 20}<|end|>
 ```
 
 Once you have gathered the output for the tool calls you can run inference with the complete content:
@@ -432,10 +432,10 @@ locations: string[],
 format?: "celsius" | "fahrenheit", // default: celsius
 }) => any;
 
-} // namespace functions<|end|><|start|>user<|message|>What is the weather like in SF?<|end|><|start|>assistant<|channel|>analysis<|message|>Need to use function get_weather.<|end|><|start|>assistant<|channel|>commentary to=functions.get_weather <|constrain|>json<|message|>{"location":"San Francisco"}<|call|><|start|>functions.get_weather to=assistant<|channel|>commentary<|message|>{"sunny": true, "temperature": 20}<|end|><|start|>assistant
+} // namespace functions<|end|><|start|>user<|message|>What is the weather like in SF?<|end|><|start|>assistant<|channel|>analysis<|message|>Need to use function get_current_weather.<|end|><|start|>assistant<|channel|>commentary to=functions.get_current_weather <|constrain|>json<|message|>{"location":"San Francisco"}<|call|><|start|>functions.get_current_weather to=assistant<|channel|>commentary<|message|>{"sunny": true, "temperature": 20}<|end|><|start|>assistant
 ```
 
-As you can see above we are passing not just the function out back into the model for further sampling but also the previous chain-of-thought (“Need to use function get_weather.”) to provide the model with the necessary information to continue its chain-of-thought or provide the final answer.
+As you can see above we are passing not just the function out back into the model for further sampling but also the previous chain-of-thought (“Need to use function get_current_weather.”) to provide the model with the necessary information to continue its chain-of-thought or provide the final answer.
 
 #### Preambles
 

examples/How_to_call_functions_with_chat_models.ipynb

@@ -51,7 +51,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "dab872c5",
    "metadata": {
     "ExecuteTime": {
@@ -66,7 +66,7 @@
     "from tenacity import retry, wait_random_exponential, stop_after_attempt\n",
     "from termcolor import colored  \n",
     "\n",
-    "GPT_MODEL = \"gpt-4o\"\n",
+    "GPT_MODEL = \"gpt-5\"\n",
     "client = OpenAI()"
    ]
   },
@@ -515,7 +515,7 @@
    "source": [
     "### Parallel Function Calling\n",
     "\n",
-    "Newer models such as gpt-4o or gpt-3.5-turbo can call multiple functions in one turn."
+    "Newer models such as gpt-5, gpt-4.1 or gpt-4o can call multiple functions in one turn."
    ]
   },
   {
@@ -758,7 +758,7 @@
    "source": [
     "##### Steps to invoke a function call using Chat Completions API: \n",
     "\n",
-    "**Step 1**: Prompt the model with content that may result in model selecting a tool to use. The description of the tools such as a function names and signature is defined in the 'Tools' list and passed to the model in API call. If selected, the function name and parameters are included in the response.<br>\n",
+    "**Step 1**: Prompt the model with content that may result in model selecting a tool to use. The description of the tools such as a function name and signature is defined in the 'Tools' list and passed to the model in API call. If selected, the function name and parameters are included in the response.<br>\n",
     "  \n",
     "**Step 2**: Check programmatically if model wanted to call a function. If true, proceed to step 3. <br>  \n",
     "**Step 3**: Extract the function name and parameters from response, call the function with parameters. Append the result to messages. <br>    \n",
@@ -767,7 +767,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": null,
    "id": "e8b7cb9cdc7a7616",
    "metadata": {
     "ExecuteTime": {
@@ -792,9 +792,9 @@
     "}]\n",
     "\n",
     "response = client.chat.completions.create(\n",
-    "    model='gpt-4o', \n",
+    "    model=GPT_MODEL, \n",
     "    messages=messages, \n",
-    "    tools= tools, \n",
+    "    tools=tools, \n",
     "    tool_choice=\"auto\"\n",
     ")\n",
     "\n",
@@ -807,7 +807,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": null,
    "id": "351c39def3417776",
    "metadata": {
     "ExecuteTime": {
@@ -847,7 +847,7 @@
     "        # Step 4: Invoke the chat completions API with the function response appended to the messages list\n",
     "        # Note that messages with role 'tool' must be a response to a preceding message with 'tool_calls'\n",
     "        model_response_with_function_call = client.chat.completions.create(\n",
-    "            model=\"gpt-4o\",\n",
+    "            model=GPT_MODEL,\n",
     "            messages=messages,\n",
     "        )  # get a new response from the model where it can see the function response\n",
     "        print(model_response_with_function_call.choices[0].message.content)\n",

registry.yaml

@@ -12,6 +12,7 @@
   tags:
     - gpt-oss
     - open-models
+    - gpt-oss-providers
 
 - title: How to run gpt-oss locally with LM Studio
   path: articles/gpt-oss/run-locally-lmstudio.md
@@ -21,6 +22,8 @@
   tags:
     - gpt-oss
     - open-models
+    - gpt-oss-local
+    
 - title: GPT-5 Prompt Migration and Improvement Using the New Optimizer
   path: examples/gpt-5/prompt-optimization-cookbook.ipynb
   date: 2025-08-07
@@ -76,6 +79,7 @@
   tags:
     - gpt-oss
     - open-models
+    - gpt-oss-server
 
 - title: Using NVIDIA TensorRT-LLM to run gpt-oss-20b
   path: articles/gpt-oss/run-nvidia.ipynb
@@ -85,6 +89,7 @@
   tags:
     - gpt-oss
     - open-models
+    - gpt-oss-server
 
 - title: Fine-tuning with gpt-oss and Hugging Face Transformers
   path: articles/gpt-oss/fine-tune-transfomers.ipynb
@@ -96,6 +101,7 @@
   tags:
     - open-models
     - gpt-oss
+    - gpt-oss-fine-tuning
 
 - title: How to handle the raw chain of thought in gpt-oss
   path: articles/gpt-oss/handle-raw-cot.md
@@ -105,6 +111,8 @@
   tags:
     - open-models
     - gpt-oss
+    - gpt-oss-fine-tuning
+    - gpt-oss-providers
 
 - title: How to run gpt-oss with Transformers
   path: articles/gpt-oss/run-transformers.md
@@ -114,6 +122,7 @@
   tags:
     - open-models
     - gpt-oss
+    - gpt-oss-server
 
 - title: How to run gpt-oss with vLLM
   path: articles/gpt-oss/run-vllm.md
@@ -123,6 +132,7 @@
   tags:
     - open-models
     - gpt-oss
+    - gpt-oss-server
 
 - title: How to run gpt-oss locally with Ollama
   path: articles/gpt-oss/run-locally-ollama.md
@@ -132,6 +142,7 @@
   tags:
     - open-models
     - gpt-oss
+    - gpt-oss-local
 
 - title: OpenAI Harmony Response Format
   path: articles/openai-harmony.md
@@ -142,6 +153,8 @@
     - open-models
     - gpt-oss
     - harmony
+    - gpt-oss-providers
+    - gpt-oss-fine-tuning
 
 - title: Temporal Agents with Knowledge Graphs
   path: examples/partners/temporal_agents_with_knowledge_graphs/temporal_agents_with_knowledge_graphs.ipynb