Patch azure embeddings + handle azure deployments properly (#594)

* Fix bug where embeddings endpoint was getting set to deployment, upgraded pinned llama-index to use new version that has azure endpoint

* updated documentation

* added memgpt example for openai

* change wording to match configure

docs/endpoints.md

@@ -8,25 +8,67 @@ set OPENAI_API_KEY=YOUR_API_KEY # on Windows
 $Env:OPENAI_API_KEY = "YOUR_API_KEY" # on Windows (PowerShell)
 ```
 
+When you run `memgpt configure`, make sure to select `openai` for both the LLM inference provider and embedding provider, for example:
+```
+$ memgpt configure
+? Select LLM inference provider: openai
+? Override default endpoint: https://api.openai.com/v1
+? Select default model (recommended: gpt-4): gpt-4
+? Select embedding provider: openai
+? Select default preset: memgpt_chat
+? Select default persona: sam_pov
+? Select default human: cs_phd
+? Select storage backend for archival data: local
+```
+
 #### OpenAI Proxies
 To use custom OpenAI endpoints, specify a proxy URL when running `memgpt configure` to set the custom endpoint as the default endpoint.
 
-
-## Azure
+## Azure OpenAI
 To use MemGPT with Azure, expore the following variables and then re-run `memgpt configure`:
 ```sh
 # see https://github.com/openai/openai-python#microsoft-azure-endpoints
-export AZURE_OPENAI_KEY = ...
-export AZURE_OPENAI_ENDPOINT = ...
-export AZURE_OPENAI_VERSION = ...
+export AZURE_OPENAI_KEY=...
+export AZURE_OPENAI_ENDPOINT=...
+export AZURE_OPENAI_VERSION=...
 
 # set the below if you are using deployment ids
-export AZURE_OPENAI_DEPLOYMENT = ...
-export AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT = ...
+export AZURE_OPENAI_DEPLOYMENT=...
+export AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT=...
+```
+
+For example, if your endpoint is `customproject.openai.azure.com` (for both your GPT model and your embeddings model), you would set the following:
+```sh
+# change AZURE_OPENAI_VERSION to the latest version
+export AZURE_OPENAI_KEY="YOUR_AZURE_KEY"
+export AZURE_OPENAI_VERSION="2023-08-01-preview"
+export AZURE_OPENAI_ENDPOINT="https://customproject.openai.azure.com"
+export AZURE_OPENAI_EMBEDDING_ENDPOINT="https://customproject.openai.azure.com"
+```
+
+If you named your deployments names other than their defaults, you would also set the following:
+```sh
+# assume you called the gpt-4 (1106-Preview) deployment "personal-gpt-4-turbo" 
+export AZURE_OPENAI_DEPLOYMENT="personal-gpt-4-turbo"
+
+# assume you called the text-embedding-ada-002 deployment "personal-embeddings" 
+export AZURE_OPENAI_EMBEDDING_DEPLOYMENT="personal-embeddings"
 ```
 
 Replace `export` with `set` or `$Env:` if you are on Windows (see the OpenAI example).
 
+When you run `memgpt configure`, make sure to select `azure` for both the LLM inference provider and embedding provider, for example:
+```
+$ memgpt configure
+? Select LLM inference provider: azure
+? Select default model (recommended: gpt-4): gpt-4-1106-preview
+? Select embedding provider: azure
+? Select default preset: memgpt_chat
+? Select default persona: sam_pov
+? Select default human: cs_phd
+? Select storage backend for archival data: local
+```
+
 Note: **your Azure endpoint must support functions** or you will get an error. See [this GitHub issue](https://github.com/cpacker/MemGPT/issues/91) for more information.
 
 ## Local Models & Custom Endpoints 

memgpt/cli/cli_config.py

@@ -19,12 +19,17 @@
 
 
 def get_azure_credentials():
-    azure_key = os.getenv("AZURE_OPENAI_KEY")
-    azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT")
-    azure_version = os.getenv("AZURE_OPENAI_VERSION")
-    azure_deployment = os.getenv("AZURE_OPENAI_DEPLOYMENT")
-    azure_embedding_deployment = os.getenv("AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT")
-    return azure_key, azure_endpoint, azure_version, azure_deployment, azure_embedding_deployment
+    creds = dict(
+        azure_key=os.getenv("AZURE_OPENAI_KEY"),
+        azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT"),
+        azure_version=os.getenv("AZURE_OPENAI_VERSION"),
+        azure_deployment=os.getenv("AZURE_OPENAI_DEPLOYMENT"),
+        azure_embedding_deployment=os.getenv("AZURE_OPENAI_EMBEDDING_DEPLOYMENT"),
+    )
+    # embedding endpoint and version default to non-embedding
+    creds["azure_embedding_endpoint"] = os.getenv("AZURE_OPENAI_EMBEDDING_ENDPOINT", creds["azure_endpoint"])
+    creds["azure_embedding_version"] = os.getenv("AZURE_OPENAI_EMBEDDING_VERSION", creds["azure_version"])
+    return creds
 
 
 def get_openai_credentials():
@@ -53,7 +58,7 @@ def configure_llm_endpoint(config: MemGPTConfig):
         provider = "openai"
     elif provider == "azure":
         model_endpoint_type = "azure"
-        _, model_endpoint, _, _, _ = get_azure_credentials()
+        model_endpoint = get_azure_credentials()["azure_endpoint"]
     else:  # local models
         backend_options = ["webui", "webui-legacy", "llamacpp", "koboldcpp", "ollama", "lmstudio", "vllm", "openai"]
         default_model_endpoint_type = None
@@ -179,7 +184,7 @@ def configure_embedding_endpoint(config: MemGPTConfig):
         embedding_dim = 1536
     elif embedding_provider == "azure":
         embedding_endpoint_type = "azure"
-        _, _, _, _, embedding_endpoint = get_azure_credentials()
+        embedding_endpoint = get_azure_credentials()["azure_embedding_endpoint"]
         embedding_dim = 1536
     elif embedding_provider == "hugging-face":
         # configure hugging face embedding endpoint (https://github.com/huggingface/text-embeddings-inference)
@@ -294,13 +299,15 @@ def configure():
     archival_storage_type, archival_storage_uri, archival_storage_path = configure_archival_storage(config)
 
     # check credentials
-    azure_key, azure_endpoint, azure_version, azure_deployment, azure_embedding_deployment = get_azure_credentials()
+    azure_creds = get_azure_credentials()
+    # azure_key, azure_endpoint, azure_version, azure_deployment, azure_embedding_deployment = get_azure_credentials()
     openai_key = get_openai_credentials()
     if model_endpoint_type == "azure" or embedding_endpoint_type == "azure":
-        if all([azure_key, azure_endpoint, azure_version]):
-            print(f"Using Microsoft endpoint {azure_endpoint}.")
-            if all([azure_deployment, azure_embedding_deployment]):
-                print(f"Using deployment id {azure_deployment}")
+        if all([azure_creds["azure_key"], azure_creds["azure_endpoint"], azure_creds["azure_version"]]):
+            print(f"Using Microsoft endpoint {azure_creds['azure_endpoint']}.")
+            # Deployment can optionally customize your endpoint model name
+            if all([azure_creds["azure_deployment"], azure_creds["azure_embedding_deployment"]]):
+                print(f"Using deployment id {azure_creds['azure_deployment']}")
         else:
             raise ValueError(
                 "Missing environment variables for Azure (see https://memgpt.readthedocs.io/en/latest/endpoints/#azure). Please set then run `memgpt configure` again."
@@ -330,11 +337,11 @@ def configure():
         agent=default_agent,
         # credentials
         openai_key=openai_key,
-        azure_key=azure_key,
-        azure_endpoint=azure_endpoint,
-        azure_version=azure_version,
-        azure_deployment=azure_deployment,
-        azure_embedding_deployment=azure_embedding_deployment,
+        azure_key=azure_creds["azure_key"],
+        azure_endpoint=azure_creds["azure_endpoint"],
+        azure_version=azure_creds["azure_version"],
+        azure_deployment=azure_creds["azure_deployment"],  # OK if None
+        azure_embedding_deployment=azure_creds["azure_embedding_deployment"],  # OK if None
         # storage
         archival_storage_type=archival_storage_type,
         archival_storage_uri=archival_storage_uri,

memgpt/embeddings.py

@@ -1,7 +1,7 @@
 import typer
 from typing import Optional, List
 import os
-from llama_index.embeddings import OpenAIEmbedding
+from llama_index.embeddings import OpenAIEmbedding, AzureOpenAIEmbedding
 from llama_index.embeddings import TextEmbeddingsInference
 from llama_index.bridge.pydantic import PrivateAttr
 
@@ -108,12 +108,11 @@ def embedding_model():
         # https://learn.microsoft.com/en-us/azure/ai-services/openai/reference#embeddings
         model = "text-embedding-ada-002"
         deployment = config.azure_embedding_deployment if config.azure_embedding_deployment is not None else model
-        return OpenAIEmbedding(
+        return AzureOpenAIEmbedding(
             model=model,
             deployment_name=deployment,
             api_key=config.azure_key,
-            api_base=config.azure_endpoint,
-            api_type="azure",
+            azure_endpoint=config.azure_endpoint,
             api_version=config.azure_version,
         )
     elif endpoint == "hugging-face":

pyproject.toml

@@ -26,7 +26,7 @@ pytz = "^2023.3.post1"
 tqdm = "^4.66.1"
 black = { version = "^23.10.1", optional = true }
 pytest = { version = "^7.4.3", optional = true }
-llama-index = "0.9.10"
+llama-index = "0.9.13"
 setuptools = "^68.2.2"
 datasets = { version = "^2.14.6", optional = true}
 prettytable = "^3.9.0"