updated ADF parameters

Author: jcodella

.env.template

@@ -15,4 +15,4 @@ FULL_TEXT_LANGUAGE=en-US
 LLM_MODEL=<your-model-deployment>
 
 ADF_ENDPOINT=http://localhost:7071/api
-ADF_KEY=
+ADF_KEY=

Samples/Demo.ipynb

@@ -43,7 +43,7 @@
     "from dotenv import load_dotenv\n",
     "from azure.identity import DefaultAzureCredential\n",
     "\n",
-    "# Add parent directory to path so we can import the package\n",
+    "# Add parent directory to path so we can import the package easily\n",
     "import sys\n",
     "sys.path.insert(0, os.path.abspath(\"..\"))\n",
     "\n",
@@ -524,6 +524,12 @@
     "print(f\"ADF key set:  {bool(memory.adf_key)}\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "b68d5b91",
+   "metadata": {},
+   "source": []
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -547,14 +553,15 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "219b6c93",
+   "id": "c2e60801",
    "metadata": {},
    "outputs": [],
    "source": [
-    "print(\"Status:\", result.get(\"runtimeStatus\"))\n",
     "print(\"\\nSummary document:\")\n",
     "output = result.get(\"output\", {})\n",
-    "print(f\"  [{output.get('thread_id', '')[:8]}...] user={output.get('user_id', ''):<10} type={output.get('type', ''):<8} {output.get('content', '')[:100]}\")"
+    "print(f\"  [{output.get('thread_id', '')[:8]}...] user={output.get('user_id', ''):<10} type={output.get('type', ''):<8} {output.get('content', '')[:100]}\")\n",
+    "\n",
+    "len(output['embedding'])"
    ]
   },
   {

azure_functions/activities.py

@@ -147,12 +147,14 @@ def generate_embeddings(payload: dict) -> list:
     Returns a list of floats (the embedding vector).
     """
     text = payload["text"]
-    model = os.environ.get("EMBEDDING_MODEL", "text-embedding-3-large")
+    model = os.environ.get("AI_FOUNDRY_EMBEDDING_MODEL", "text-embedding-3-large")
 
+    dimensions = int(os.environ.get("EMBEDDING_DIMENSION", "1536"))
     client = _get_embeddings_client()
     response = client.embeddings.create(
         input=[text],
         model=model,
+        dimensions=dimensions,
     )
     return response.data[0].embedding
 
@@ -222,7 +224,7 @@ def generate_thread_summary(payload: dict) -> dict:
     user_id = payload["user_id"]
     thread_id = payload["thread_id"]
     recent_k = payload.get("recent_k")
-    model = os.environ.get("LLM_MODEL", "gpt-4o")
+    model = os.environ.get("LLM_MODEL", "gpt-5-nano")
     container = _get_cosmos_container()
 
     # ---- 1. Check for an existing thread summary ----
@@ -309,11 +311,13 @@ def generate_thread_summary(payload: dict) -> dict:
     summary_text = response.choices[0].message.content
 
     # ---- 6. Generate embedding ----
-    embedding_model = os.environ.get("EMBEDDING_MODEL", "text-embedding-3-large")
+    embedding_model = os.environ.get("AI_FOUNDRY_EMBEDDING_MODEL", "text-embedding-3-large")
+    dimensions = int(os.environ.get("EMBEDDING_DIMENSION", "1536"))
     emb_client = _get_embeddings_client()
     emb_response = emb_client.embeddings.create(
         input=[summary_text],
         model=embedding_model,
+        dimensions=dimensions,
     )
     summary_embedding = emb_response.data[0].embedding
 
@@ -370,7 +374,7 @@ def extract_facts(payload: dict) -> dict:
     user_id = payload["user_id"]
     thread_id = payload["thread_id"]
     recent_k = payload.get("recent_k")
-    model = os.environ.get("LLM_MODEL", "gpt-4o")
+    model = os.environ.get("AI_FOUNDRY_LLM", "gpt-5-nano")
 
     # ---- 1. Query Cosmos DB ----
     container = _get_cosmos_container()
@@ -434,7 +438,8 @@ def extract_facts(payload: dict) -> dict:
         fact_lines = [facts_text.strip()]
 
     # ---- 5. Generate embeddings and store each fact ----
-    embedding_model = os.environ.get("EMBEDDING_MODEL", "text-embedding-3-large")
+    embedding_model = os.environ.get("AI_FOUNDRY_EMBEDDING_MODEL", "text-embedding-3-large")
+    dimensions = int(os.environ.get("EMBEDDING_DIMENSION", "1536"))
     emb_client = _get_embeddings_client()
     now = datetime.now(timezone.utc).isoformat()
     facts_docs = []
@@ -443,6 +448,7 @@ def extract_facts(payload: dict) -> dict:
         emb_response = emb_client.embeddings.create(
             input=[fact],
             model=embedding_model,
+            dimensions=dimensions,
         )
         fact_doc = {
             "id": str(uuid.uuid4()),
@@ -490,7 +496,7 @@ def generate_user_summary(payload: dict) -> dict:
     user_id = payload["user_id"]
     thread_ids = payload.get("thread_ids")
     recent_k = payload.get("recent_k")
-    model = os.environ.get("LLM_MODEL", "gpt-4o")
+    model = os.environ.get("AI_FOUNDRY_LLM", "gpt-5-nano")
     container = _get_cosmos_container()
 
     # ---- 1. Check for an existing user summary ----
@@ -598,11 +604,13 @@ def generate_user_summary(payload: dict) -> dict:
     summary_text = response.choices[0].message.content
 
     # ---- 6. Generate embedding ----
-    embedding_model = os.environ.get("EMBEDDING_MODEL", "text-embedding-3-large")
+    embedding_model = os.environ.get("AI_FOUNDRY_EMBEDDING_MODEL", "text-embedding-3-large")
+    dimensions = int(os.environ.get("EMBEDDING_DIMENSION", "1536"))
     emb_client = _get_embeddings_client()
     emb_response = emb_client.embeddings.create(
         input=[summary_text],
         model=embedding_model,
+        dimensions=dimensions,
     )
     summary_embedding = emb_response.data[0].embedding
 

azure_functions/local.settings.json.template

@@ -3,12 +3,18 @@
   "Values": {
     "AzureWebJobsStorage": "UseDevelopmentStorage=true",
     "FUNCTIONS_WORKER_RUNTIME": "python",
-    "COSMOS_DB_ENDPOINT": "https://<your-account>.documents.azure.com:443/",
+    "languageWorkers:python:defaultExecutablePath": "/Users/jcodella/github/AgentMemoryToolkit/azure_functions/.venv/bin/python",
+
+    "COSMOS_DB_ENDPOINT": "https://<cosmos-endpoint>.documents.azure.com:443/",
     "COSMOS_DB_DATABASE": "ai_memory",
     "COSMOS_DB_CONTAINER": "memories",
-    "AI_FOUNDRY_ENDPOINT": "https://<your-project>.services.ai.azure.com/",
-    "EMBEDDING_MODEL": "text-embedding-3-large",
-    "EMBEDDING_DIMENSIONS": "1536",
-    "LLM_MODEL": "gpt-5-mini"
+
+    "SUMMARIZE_EVERY_N_MESSAGES": "0",
+
+    "AI_FOUNDRY_ENDPOINT": "https://<foundry-endpoint>.openai.azure.com/",
+    "AI_FOUNDRY_LLM": "gpt-5-nano",
+    "AI_FOUNDRY_EMBEDDING_MODEL": "text-embedding-3-large",
+    "EMBEDDING_DIMENSION": "1536",
+    "AI_FOUNDRY_API_VERSION": "2024-12-01-preview"
   }
 }