Reworking the LLM layer (#666)

deployment/docker_compose/docker-compose.dev.yml

@@ -16,11 +16,11 @@ services:
     ports:
       - "8080:8080"
     environment:
-      - INTERNAL_MODEL_VERSION=${INTERNAL_MODEL_VERSION:-openai-chat-completion}
+      - GEN_AI_MODEL_PROVIDER=${GEN_AI_MODEL_PROVIDER:-openai}
       - GEN_AI_MODEL_VERSION=${GEN_AI_MODEL_VERSION:-gpt-3.5-turbo}
       - GEN_AI_API_KEY=${GEN_AI_API_KEY:-}
-      - GEN_AI_ENDPOINT=${GEN_AI_ENDPOINT:-}
-      - GEN_AI_HOST_TYPE=${GEN_AI_HOST_TYPE:-}
+      - GEN_AI_API_ENDPOINT=${GEN_AI_API_ENDPOINT:-}
+      - GEN_AI_API_VERSION=${GEN_AI_API_VERSION:-}
       - NUM_DOCUMENT_TOKENS_FED_TO_GENERATIVE_MODEL=${NUM_DOCUMENT_TOKENS_FED_TO_GENERATIVE_MODEL:-}
       - POSTGRES_HOST=relational_db
       - VESPA_HOST=index
@@ -30,10 +30,6 @@ services:
       - GOOGLE_OAUTH_CLIENT_ID=${GOOGLE_OAUTH_CLIENT_ID:-}
       - GOOGLE_OAUTH_CLIENT_SECRET=${GOOGLE_OAUTH_CLIENT_SECRET:-}
       - DISABLE_GENERATIVE_AI=${DISABLE_GENERATIVE_AI:-}
-      - API_BASE_OPENAI=${API_BASE_OPENAI:-}
-      - API_TYPE_OPENAI=${API_TYPE_OPENAI:-}
-      - API_VERSION_OPENAI=${API_VERSION_OPENAI:-}
-      - AZURE_DEPLOYMENT_ID=${AZURE_DEPLOYMENT_ID:-}
       - NOTION_CONNECTOR_ENABLE_RECURSIVE_PAGE_LOOKUP=${NOTION_CONNECTOR_ENABLE_RECURSIVE_PAGE_LOOKUP:-}
       - DISABLE_TIME_FILTER_EXTRACTION=${DISABLE_TIME_FILTER_EXTRACTION:-}
       # Don't change the NLP model configs unless you know what you're doing
@@ -63,17 +59,13 @@ services:
       - index
     restart: always
     environment:
-      - INTERNAL_MODEL_VERSION=${INTERNAL_MODEL_VERSION:-openai-chat-completion}
+      - GEN_AI_MODEL_PROVIDER=${GEN_AI_MODEL_PROVIDER:-openai}
       - GEN_AI_MODEL_VERSION=${GEN_AI_MODEL_VERSION:-gpt-3.5-turbo}
       - GEN_AI_API_KEY=${GEN_AI_API_KEY:-}
-      - GEN_AI_ENDPOINT=${GEN_AI_ENDPOINT:-}
-      - GEN_AI_HOST_TYPE=${GEN_AI_HOST_TYPE:-}
+      - GEN_AI_API_ENDPOINT=${GEN_AI_API_ENDPOINT:-}
+      - GEN_AI_API_VERSION=${GEN_AI_API_VERSION:-}
       - POSTGRES_HOST=relational_db
       - VESPA_HOST=index
-      - API_BASE_OPENAI=${API_BASE_OPENAI:-}
-      - API_TYPE_OPENAI=${API_TYPE_OPENAI:-}
-      - API_VERSION_OPENAI=${API_VERSION_OPENAI:-}
-      - AZURE_DEPLOYMENT_ID=${AZURE_DEPLOYMENT_ID:-}
       - NUM_INDEXING_WORKERS=${NUM_INDEXING_WORKERS:-}
       # Connector Configs
       - CONTINUE_ON_CONNECTOR_FAILURE=${CONTINUE_ON_CONNECTOR_FAILURE:-}

deployment/docker_compose/env.prod.template

@@ -3,57 +3,57 @@
 # This is only necessary when using the docker-compose.prod.yml compose file.
 
 
-# Insert your OpenAI API key here If not provided here, UI will prompt on setup.
-# This env variable takes precedence over UI settings.
-GEN_AI_API_KEY=
-# Choose between "openai-chat-completion" and "openai-completion"
-INTERNAL_MODEL_VERSION=openai-chat-completion
-# Use a valid model for the choice above, consult https://platform.openai.com/docs/models/model-endpoint-compatibility
-GEN_AI_MODEL_VERSION=gpt-4
-
-# Neccessary environment variables for Azure OpenAI:
-API_BASE_OPENAI=
-API_TYPE_OPENAI=
-API_VERSION_OPENAI=
-AZURE_DEPLOYMENT_ID=
-
 # Could be something like danswer.companyname.com
 WEB_DOMAIN=http://localhost:3000
 
-# Default values here are what Postgres uses by default, feel free to change.
-POSTGRES_USER=postgres
-POSTGRES_PASSWORD=password
+
+# Generative AI settings, uncomment as needed, will work with defaults
+GEN_AI_MODEL_PROVIDER=openai
+GEN_AI_MODEL_VERSION=gpt-4
+# Provide this as a global default/backup, this can also be set via the UI
+#GEN_AI_API_KEY=
+# Set to use Azure OpenAI or other services, such as https://danswer.openai.azure.com/
+#GEN_AI_API_ENDPOINT=
+# Set up to use a specific API version, such as 2023-09-15-preview (example taken from Azure)
+#GEN_AI_API_VERSION=
+
 
 # If you want to setup a slack bot to answer questions automatically in Slack 
-# channels it is added to, you must specify the below. 
+# channels it is added to, you must specify the two below.
 # More information in the guide here: https://docs.danswer.dev/slack_bot_setup
-DANSWER_BOT_SLACK_APP_TOKEN=
-DANSWER_BOT_SLACK_BOT_TOKEN=
+#DANSWER_BOT_SLACK_APP_TOKEN=
+#DANSWER_BOT_SLACK_BOT_TOKEN=
 
-# Used to generate values for security verification, use a random string
-SECRET=
-
-# How long before user needs to reauthenticate, default to 1 day. (cookie expiration time)
-SESSION_EXPIRE_TIME_SECONDS=86400
 
 # The following are for configuring User Authentication, supported flows are:
 # disabled
 # google_oauth (login with google/gmail account)
 # oidc (only in Danswer enterprise edition)
 # saml (only in Danswer enterprise edition)
-AUTH_TYPE=
+AUTH_TYPE=google_oauth
 
-# Set the two below to use with Google OAuth
+# Set the values below to use with Google OAuth
 GOOGLE_OAUTH_CLIENT_ID=
 GOOGLE_OAUTH_CLIENT_SECRET=
+SECRET=
 
 # OpenID Connect (OIDC)
-OPENID_CONFIG_URL=
+#OPENID_CONFIG_URL=
 
 # SAML config directory for OneLogin compatible setups
-SAML_CONF_DIR=
+#SAML_CONF_DIR=
 
-# used to specify a list of allowed user domains, only checked if user Auth is turned on
+
+# How long before user needs to reauthenticate, default to 1 day. (cookie expiration time)
+SESSION_EXPIRE_TIME_SECONDS=86400
+
+
+# Use the below to specify a list of allowed user domains, only checked if user Auth is turned on
 # e.g. `VALID_EMAIL_DOMAINS=example.com,example.org` will only allow users
 # with an @example.com or an @example.org email
-VALID_EMAIL_DOMAINS=
+#VALID_EMAIL_DOMAINS=
+
+
+# Default values here are what Postgres uses by default, feel free to change.
+POSTGRES_USER=postgres
+POSTGRES_PASSWORD=password