init commit

.github/workflows/deploy.yml

@@ -0,0 +1,48 @@
+name: docusaurus.io
+
+on:
+  push:
+    branches:
+     # - gh-pages
+      - main
+    # Review gh actions docs if you want to further define triggers, paths, etc
+    # https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#on
+
+permissions:
+  contents: write
+
+jobs:
+  deploy:
+    name: Deploy to GitHub Pages
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-node@v3
+        with:
+          node-version: 18
+          cache: npm
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Install docusaurus
+        run: npm install --global docusaurus-init
+
+      - name: Build website
+        run: npm run build
+
+      # Popular action to deploy to GitHub Pages:
+      # Docs: https://github.com/peaceiris/actions-gh-pages#%EF%B8%8F-docusaurus
+      - name: Deploy to GitHub Pages
+        uses: peaceiris/actions-gh-pages@v3
+        with:
+          github_token: ${{ secrets.GITHUB_TOKEN }}
+          # Build output to publish to the `gh-pages` branch:
+          publish_dir: ./build
+          # The following lines assign commit authorship to the official
+          # GH-Actions bot for deploys to `gh-pages` branch:
+          # https://github.com/actions/checkout/issues/13#issuecomment-724415212
+          # The GH actions bot is used by default if you didn't specify the two fields.
+          # You can swap them out with your own user credentials.
+          user_name: github-actions[bot]
+          user_email: 41898282+github-actions[bot]@users.noreply.github.com

.github/workflows/test-deploy.yml

@@ -0,0 +1,29 @@
+name: Test deployment
+
+on:
+  pull_request:
+    branches:
+      - main
+    # Review gh actions docs if you want to further define triggers, paths, etc
+    # https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#on
+
+jobs:
+  test-deploy:
+    name: Test deployment
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-node@v3
+        with:
+          node-version: 18
+          cache: npm
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Install docusaurus
+        run: npm install --global docusaurus-init
+
+      - name: Build website
+        run: npm run build
+

.gitignore

@@ -0,0 +1,20 @@
+# Dependencies
+/node_modules
+
+# Production
+/build
+
+# Generated files
+.docusaurus
+.cache-loader
+
+# Misc
+.DS_Store
+.env.local
+.env.development.local
+.env.test.local
+.env.production.local
+
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*

babel.config.js

@@ -0,0 +1,3 @@
+module.exports = {
+  presets: [require.resolve('@docusaurus/core/lib/babel/preset')],
+};

docs/Deployment/Azure Model Deployment.md

@@ -0,0 +1,30 @@
+## Create and Deploy OpenAI Model in Azure
+
+In this section, you will learn how to create an Azure OpenAI model GPT-3.5-turbo/GPT-4.
+
+1. Login to the Azure portal (https://portal.azure.com/)
+2. Use the search bar to look up **Azure OpenAI** and click it to navigate to the **Azure AI Services|Azure OpenAI** page.
+
+   ![](img/step2.jpg)
+
+3. In **Azure AI Services|Azure OpenAI**, click **Create** and fill in all the required fields.
+
+   ![](img/step3.jpg)
+
+4. You may need to request access to Azure OpenAI Services. Follow the link in the notification to do that.
+
+   ![](img/step5.jpg)
+
+5. When done, you should have your OpenAI model in the **Azure AI services** section. Click it to open and then click **Go to Azure OpenAI Studio** in the top bar.
+
+   ![](img/step8.jpg)
+
+6. In Azure OpenAI Studio, click **Deployment** in the navigation menu and click **Create new deployment**. Fill in the required fields and click **Create** to create a model.
+
+   ![](img/step9.jpg)
+
+    > It is important to note that certain models may not be accessible for deployment in a particular region. If you need a particular model, you will have to submit a separate request or relocate Azure OpenAI to a different region.
+
+7. Go back to your model page and click **Keys and Endpoint**. In this section, you can find your key and endpoint that you will need to provide in [AI DIAL configuration file](./dialConfig.yaml#L30).
+
+    ![](img/step13.jpg)

docs/Deployment/configuration.md

@@ -0,0 +1,98 @@
+# Configuration
+
+> Refer to the provided [example of config](./dialConfig.yaml), where you can find the description of the application-specific parameters.
+
+The `dialConfig.yaml` configuration file of the AI DIAL application is comprised of several main sections:
+
+* Standard parameters of helm chart. Refer to [Helm docs](https://helm.sh/).
+* [Front-end parameters](#front-end-parameters): in `env` and `secrets` sections in config.
+* [Back-end parameters](#back-end-parameters): in `proxy` section in the config.
+* [Configuration of Adapters for models](#configuration-of-adapters). 
+
+> **Important**: it is assumed that you have a working knowledge of standard Helm chart parameters in order to define them within the configuration file.
+
+## Front-End Parameters
+
+> Refer to the [AI DIAL Chat](https://github.com/epam/ai-dial-chat#environment-variables) to view a complete documentation.
+
+Configure front-end parameters in the [`env`](./dialConfig.yaml#L18) and [`secrets`](./dialConfig.yaml#L30) sections of the config file:
+
+|Parameter|Description|
+|---------|-----------|
+|NEXTAUTH_URL|Public URL of the application. When deploying to production, set the `NEXTAUTH_URL` environment variable to the canonical URL of your site.|
+|NEXTAUTH_SECRET|A random string used as a ceed for authentication. Used to encrypt the NextAuth.js JWT, and to hash email verification tokens.|
+|NEXT_PUBLIC_DEFAULT_TEMPERATURE|Default temperature settings in the range: [0 1]|
+|NEXT_PUBLIC_DEFAULT_SYSTEM_PROMPT|The default system prompt.|
+|NEXT_PUBLIC_APP_NAME | Public Application Name |
+|OPENAI_API_HOST|AI DIAL back-end URL.|
+|OPENAI_API_KEY|Open AI API Key.|
+|OPENAI_API_VERSION|Version of the OpenAI API.|
+|DEFAULT_MODEL|Default LLM.|
+|ENABLED_FEATURES|A list of UI features.|
+|NEXT_PUBLIC_APP_NAME|Application name.|
+|AVAILABLE_MODELS_USERS_LIMITATIONS|Specify models and users that have access to them. Skip to allow all users to access all models.|
+|AVAILABLE_ADDONS_USERS_LIMITATIONS|Specify Addons and users that have access to them. Skip to allow all users to access all Addons.|
+|CLIENT_ID|You client id at auth provider.|
+|TENANT_ID|You tenant id at auth provider.|
+|SECRET|You secret at auth provider.|
+|NAME|Display name in AI DIAL app.|
+|HOST|Auth provider URL.|
+|AUDIENCE|Your audience at auth provider.|
+
+## Back-End Parameters
+
+> Refer to the [AI DIAL Core](https://github.com/epam/ai-dial-core) to view a complete documentation.
+
+Configure back-end parameters in the [`proxy`](./dialConfig.yaml#L74) section of the config file.
+
+You can provide dynamic or static configurations for the back-end. Provide path to the corresponding configuration in the `proxy.env` section.
+
+Static settings are configured at application startup and do not change throughout the application lifecycle. 
+
+Priority order of configurations:
+
+* Environment variables with extra "proxy." prefix. E.g. "proxy.server.port", "proxy.config.files".
+* File specified in "PROXY_SETTINGS" environment variable.
+* Default resource file: src/main/resources/proxy.settings.json.
+
+|Parameter|Default Value|Description|
+|---------|-----------|-------------|
+|config.files|proxy.config.json |Config files with parts of the whole config.|
+|config.reload|60000|Config reload interval in milliseconds.|
+|identityProvider.jwksUrl|-|URL to the jwks provider.|
+|identityProvider.appName|dial|App name to search in "resource_access" claim of JWT token to check access for deployments.|
+|identityProvider.loggingKey|-|User information to search in claims of JWT token.|
+|identityProvider.loggingSalt|-|Salt to hash user information for logging.|
+|identityProvider.cacheSize|10|How many JWT tokens to cache.|
+|identityProvider.cacheExpiration|10|How long to retain JWT tokens in cache.|
+|identityProvider.cacheExpirationUnit|MINUTES|Unit of cache expiration.|
+
+This file includes standard [Vertex library configurations](https://cloud.google.com/vertex-ai/docs/start/client-libraries).
+
+**Dynamic** settings are defined in the `proxy.config.json` file by default. You can override settings in this file by configuring static settings.
+
+### proxy.config.json parameters
+
+> Refer to the [confuration file](./dialConfig.yaml#L112) to view an example.
+
+|Parameter|Description|
+|---------|-----------|
+|routes|Path(s)  Paths to route to specific upstreams or to respond with configured body.|
+|route-rate|Parameters for vote endpoint.|
+|applications|A list of deployed AI DIAL Applications and their parameters:<br />`endpoint`: AI DIAL Application API for chat completions.<br />`iconUrl`: a path to the icon used for the AI DIAL Application on the UI.<br />`description`: a brief description of the AI DIAL Application rendered on the UI.<br />`displayName`: a name of the AI DIAL Application used on the UI.|
+|models|A list of deployed models and their parameters:<br />`type`: specify `chat` or `embedding` model type.<br />`iconUrl`: a path to the icon used for the model on the UI.<br />`description`: a brief description of the model rendered on the UI.<br />`displayName`: a name of the model rendered on the UI.<br />`endpoint`: model API for chat completions or embeddings.<br />`upstreams`: upstreams are used for load-balancing. A request will be sent to the configured model endpoint and will contain X-UPSTREAM-ENDPOINT and X-UPSTREAM-KEY headers:<br />`endpoint`: model endpoint.<br />`key`: your API key.|
+|keys|API Keys parameters:<br />`<proxyKey>`: your API key.<br />`project`: a project name this key is assigned to.<br />`role`: name of one of the configured roles. Defines permissions for the key.<br />`userAuth`: can be disabled/enabled/optional.<br />**Disabled** - Authorization header is ignored and not sent to upstream.<br />**Enabled** - Authorization header is required and sent to upstream. Optional - Authorization header is optional and sent to upstream if present.|
+|roles|A list of configured roles with their limitations. Specify a role name and in limits, specify models (can be models, Applications, Addons, Assistants) and limitation configurations:<br />`minute`: the total tokens per minute limit that can be sent to the model is managed using a floating window approach. This technique ensures a well-distributed rate-limiting mechanism, allowing control over the number of tokens sent to the model within a defined time frame, typically a one-minute window.<br />`day`: the total tokens per day limit that can be sent to the model is managed using a floating window approach. This method ensures a balanced rate-limiting mechanism, allowing control over the number of tokens sent to the model within a specified time frame, typically a 24-hour window.|
+
+## Configuration of Adapters 
+
+To work with Azure, AWS or GCP models we use applications called Adapters. You can configure Adapters in the configuration file.
+
+Refer to these repositories to view a complete documentation for: 
+
+* [Adapter for Bedrock](https://github.com/epam/ai-dial-adapter-bedrock)
+* [Adapter for Vertex](https://github.com/epam/ai-dial-adapter-vertexai)
+* [Adapter for OpenAI](https://github.com/epam/ai-dial-adapter-openai)
+
+> Refer to the provided [example of config](./dialConfig.yaml#L263) to view configuration examples. 
+