add AzureOpenAI Embedding Model Support

.env

@@ -1,2 +1,22 @@
+OPENAI_API_TYPE=placeholder
+# if OPENAI_API_TYPE is "azure", then the OPENAI_API_KEY is the Azure API key
 OPENAI_API_KEY=placeholder
+
+AZURE_OPENAI_API_BASE=placeholder
+AZURE_OPENAI_API_VERSION=placeholder
+AZURE_OPENAI_API_KEY=placeholder
+AZURE_OPENAI_DEPLOYMENT_NAME=placeholder
+AZURE_OPENAI_DEPLOYMENT_NAME_EB="your_azure_embedding_deployment_name_here"
+
+REDIS_URL=placeholder
+
 ANTHROPIC_API_KEY=placeholder
+YDC_API_KEY=your_secret_here
+LANGCHAIN_TRACING_V2="true"
+LANGCHAIN_PROJECT=langserve-launch-example
+LANGCHAIN_API_KEY=your_secret_key_here
+TAVILY_API_KEY=your_secret_here
+FIREWORKS_API_KEY=your_secret_here
+AWS_ACCESS_KEY_ID=your_secret_here
+AWS_SECRET_ACCESS_KEY=your_secret_here
+KAY_API_KEY=your_secret_here

backend/packages/gizmo-agent/gizmo_agent/ingest.py

@@ -9,16 +9,28 @@
 index_schema = {
     "tag": [{"name": "namespace"}],
 }
+
+if os.environ["OPENAI_API_TYPE"] == "azure":
+    embeddings = OpenAIEmbeddings(
+        deployment=os.environ["AZURE_OPENAI_DEPLOYMENT_NAME_EB"],
+        model=os.environ["AZURE_OPENAI_DEPLOYMENT_NAME_EB"],
+        openai_api_base=os.environ["AZURE_OPENAI_API_BASE"],
+        openai_api_type="azure",
+    )
+else:
+    embeddings = OpenAIEmbeddings()
+
 vstore = Redis(
     redis_url=os.environ["REDIS_URL"],
     index_name="opengpts",
-    embedding=OpenAIEmbeddings(),
+    embedding=embeddings,
     index_schema=index_schema,
 )
 
 
 ingest_runnable = IngestRunnable(
-    text_splitter=RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200),
+    text_splitter=RecursiveCharacterTextSplitter(
+        chunk_size=1000, chunk_overlap=200),
     vectorstore=vstore,
 ).configurable_fields(
     assistant_id=ConfigurableField(