Merge branch 'master' into patch-1

.env.template

@@ -17,6 +17,10 @@ OPENAI_AZURE_API_BASE=your-base-url-for-azure
 OPENAI_AZURE_API_VERSION=api-version-for-azure
 # OPENAI_AZURE_DEPLOYMENT_ID - OpenAI deployment ID for Azure (Example: my-deployment-id)
 OPENAI_AZURE_DEPLOYMENT_ID=deployment-id-for-azure
+# OPENAI_AZURE_CHAT_DEPLOYMENT_ID - OpenAI deployment ID for Azure Chat (Example: my-deployment-id-for-azure-chat)
+OPENAI_AZURE_CHAT_DEPLOYMENT_ID=deployment-id-for-azure-chat
+# OPENAI_AZURE_EMBEDDINGS_DEPLOYMENT_ID - OpenAI deployment ID for Embedding (Example: my-deployment-id-for-azure-embeddigs)
+OPENAI_AZURE_EMBEDDINGS_DEPLOYMENT_ID=deployment-id-for-azure-embeddigs
 
 ################################################################################
 ### LLM MODELS
@@ -39,7 +43,7 @@ SMART_TOKEN_LIMIT=8000
 ################################################################################
 
 # MEMORY_BACKEND - Memory backend type (Default: local)
-MEMORY_BACKEND=redis
+MEMORY_BACKEND=local
 
 ### PINECONE
 # PINECONE_API_KEY - Pinecone API Key (Example: my-pinecone-api-key)

README.md

@@ -93,7 +93,7 @@ pip install -r requirements.txt
 4. Rename `.env.template` to `.env` and fill in your `OPENAI_API_KEY`. If you plan to use Speech Mode, fill in your `ELEVEN_LABS_API_KEY` as well.
   - Obtain your OpenAI API key from: https://platform.openai.com/account/api-keys.
   - Obtain your ElevenLabs API key from: https://elevenlabs.io. You can view your xi-api-key using the "Profile" tab on the website.
-  - If you want to use GPT on an Azure instance, set `USE_AZURE` to `True` and provide the `OPENAI_AZURE_API_BASE`, `OPENAI_AZURE_API_VERSION` and `OPENAI_AZURE_DEPLOYMENT_ID` values as explained here: https://pypi.org/project/openai/ in the `Microsoft Azure Endpoints` section
+  - If you want to use GPT on an Azure instance, set `USE_AZURE` to `True` and provide the `OPENAI_AZURE_API_BASE`, `OPENAI_AZURE_API_VERSION` and `OPENAI_AZURE_DEPLOYMENT_ID` values as explained here: https://pypi.org/project/openai/ in the `Microsoft Azure Endpoints` section. Additionally you need separate deployments for both embeddings and chat. Add their ID values to `OPENAI_AZURE_CHAT_DEPLOYMENT_ID` and `OPENAI_AZURE_EMBEDDINGS_DEPLOYMENT_ID` respectively
 
 ## 🔧 Usage
 

scripts/config.py

@@ -49,6 +49,8 @@ class Config(metaclass=Singleton):
             self.openai_api_base = os.getenv("OPENAI_AZURE_API_BASE")
             self.openai_api_version = os.getenv("OPENAI_AZURE_API_VERSION")
             self.openai_deployment_id = os.getenv("OPENAI_AZURE_DEPLOYMENT_ID")
+            self.azure_chat_deployment_id = os.getenv("OPENAI_AZURE_CHAT_DEPLOYMENT_ID")
+            self.azure_embeddigs_deployment_id = os.getenv("OPENAI_AZURE_EMBEDDINGS_DEPLOYMENT_ID")
             openai.api_type = "azure"
             openai.api_base = self.openai_api_base
             openai.api_version = self.openai_api_version

scripts/json_parser.py

@@ -26,7 +26,7 @@ JSON_SCHEMA = """
 """
 
 
-def fix_and_parse_json(    
+def fix_and_parse_json(
     json_str: str,
     try_to_fix_with_gpt: bool = True
 ) -> Union[str, Dict[Any, Any]]:
@@ -35,8 +35,8 @@ def fix_and_parse_json(
         json_str = json_str.replace('\t', '')
         return json.loads(json_str)
     except json.JSONDecodeError as _:  # noqa: F841
-        json_str = correct_json(json_str)
         try:
+            json_str = correct_json(json_str)
             return json.loads(json_str)
         except json.JSONDecodeError as _:  # noqa: F841
             pass
@@ -53,6 +53,7 @@ def fix_and_parse_json(
         last_brace_index = json_str.rindex("}")
         json_str = json_str[:last_brace_index+1]
         return json.loads(json_str)
+    # Can throw a ValueError if there is no "{" or "}" in the json_str
     except (json.JSONDecodeError, ValueError) as e:  # noqa: F841
         if try_to_fix_with_gpt:
             print("Warning: Failed to parse AI output, attempting to fix."

scripts/llm_utils.py

@@ -9,7 +9,7 @@ def create_chat_completion(messages, model=None, temperature=None, max_tokens=No
     """Create a chat completion using the OpenAI API"""
     if cfg.use_azure:
         response = openai.ChatCompletion.create(
-            deployment_id=cfg.openai_deployment_id,
+            deployment_id=cfg.azure_chat_deployment_id,
             model=model,
             messages=messages,
             temperature=temperature,

scripts/memory/base.py

@@ -1,12 +1,16 @@
 """Base class for memory providers."""
 import abc
-from config import AbstractSingleton
+from config import AbstractSingleton, Config
 import openai
+cfg = Config()
 
 
 def get_ada_embedding(text):
     text = text.replace("\n", " ")
-    return openai.Embedding.create(input=[text], model="text-embedding-ada-002")["data"][0]["embedding"]
+    if cfg.use_azure:
+        return openai.Embedding.create(input=[text], engine=cfg.azure_embeddigs_deployment_id, model="text-embedding-ada-002")["data"][0]["embedding"]
+    else:
+        return openai.Embedding.create(input=[text], model="text-embedding-ada-002")["data"][0]["embedding"]
 
 
 class MemoryProviderSingleton(AbstractSingleton):