Merge pull request #208 from Azure-Samples/model-router

Adding gpt-5

Author: nourshaker-msft

images/model-routing.gif

@@ -58,6 +58,7 @@
     "                 {\"name\": \"gpt-4.1-mini\", \"publisher\": \"OpenAI\", \"version\": \"2025-04-14\", \"sku\": \"GlobalStandard\", \"capacity\": 20, \"aiservice\": \"foundry2\"},\n",
     "                 {\"name\": \"gpt-4.1-nano\", \"publisher\": \"OpenAI\", \"version\": \"2025-04-14\", \"sku\": \"GlobalStandard\", \"capacity\": 20, \"aiservice\": \"foundry2\"},\n",
     "                 {\"name\": \"model-router\", \"publisher\": \"OpenAI\", \"version\": \"2025-05-19\", \"sku\": \"GlobalStandard\", \"capacity\": 20, \"aiservice\": \"foundry3\"},\n",
+    "                 {\"name\": \"gpt-5\", \"publisher\": \"OpenAI\", \"version\": \"2025-08-07\", \"sku\": \"GlobalStandard\", \"capacity\": 20, \"aiservice\": \"foundry3\"},\n",
     "                 {\"name\": \"DeepSeek-R1\", \"publisher\": \"DeepSeek\", \"version\": \"1\", \"sku\": \"GlobalStandard\", \"capacity\": 20, \"aiservice\": \"foundry3\"}]\n",
     "\n",
     "apim_sku = 'Basicv2'\n",
@@ -202,7 +203,7 @@
     "    api_version=inference_api_version\n",
     ")\n",
     "try:\n",
-    "    for model in ['model-router', 'DeepSeek-R1', 'gpt-4.1']:\n",
+    "    for model in ['model-router', 'DeepSeek-R1', 'gpt-4.1', 'gpt-5']:\n",
     "        completion = client.chat.completions.with_raw_response.create(model=model, messages=messages)\n",
     "        # print(\"headers \", completion.headers)\n",
     "        print(\"x-ms-region: \", completion.headers.get(\"x-ms-region\")) # this header is useful to determine the region of the backend that served the request\n",
@@ -238,7 +239,7 @@
     "    api_version=inference_api_version\n",
     ")\n",
     "try:\n",
-    "    for model in ['gpt-4.1-mini', 'gpt-4.1-nano', 'gpt-4.1']:\n",
+    "    for model in ['gpt-4.1-mini', 'gpt-4.1-nano', 'gpt-4.1', 'gpt-5']:\n",
     "        responses = client.responses.with_raw_response.create(model=model, input=input_message)\n",
     "        # print(\"headers \", responses.headers)\n",
     "        print(\"x-ms-region: \", responses.headers.get(\"x-ms-region\"))\n",

labs/model-routing/model-routing.ipynb

@@ -27,7 +27,8 @@
                 <set-backend-service backend-id="foundry2" />
             </when>
             <when condition="@( ((string)context.Variables["requestedModel"]) == "model-router"
-                         || ((string)context.Variables["requestedModel"]) == "DeepSeek-R1")">
+                            || ((string)context.Variables["requestedModel"]) == "gpt-5"
+                            || ((string)context.Variables["requestedModel"]) == "DeepSeek-R1")">
                 <set-backend-service backend-id="foundry3" />
             </when>
             <!-- gate any GPT-4o* variants -->