Red Hat Chat - AI Conversations with Red Hat Design System

Red Hat Chat - Deploy locally with Podman or on OpenShift

📋 TL;DR

Local Deployment: Run with Podman using podman-compose or podman compose
OpenShift Deployment: Install from the catalog using image quay.io/maximilianopizarro/redhat-chat:latest
Container Image: Available on Quay.io
Requirements: Podman 4.0+ for local, OpenShift 4.x+ for cluster deployment

📖 Overview

Red Hat Chat brings together the future of assistant AIs with the revolutionary technology of OpenAI’s ChatGPT, styled with the Red Hat Design System. This application integrates multiple AI models while maintaining Red Hat’s design standards and visual identity.

Key Features

🖥️ UI & Experience powered by Red Hat Design System with enhanced design and features
🤖 AI Model Selection: Anthropic (Claude), AWS Bedrock, OpenAI, Azure OpenAI, Google, Vertex AI, and more
🔧 Code Interpreter API: Secure, sandboxed execution in Python, Node.js, Go, C/C++, Java, PHP, Rust, and Fortran
🔦 Agents & Tools Integration: No-code custom assistants with MCP server support
🔍 Web Search: Search the internet and retrieve relevant information
🎨 Image Generation & Editing: Text-to-image and image-to-image capabilities
💬 Multimodal & File Interactions: Upload and analyze images, chat with files
🌎 Multilingual UI: Support for 20+ languages
👥 Multi-User & Secure Access: OAuth2, LDAP, & Email Login Support

Container Image

The application is available as a container image on Quay.io:

Repository: quay.io/maximilianopizarro/redhat-chat
Latest Tag: latest
Base Image: Red Hat UBI8 with Node.js 20
Architecture: Multi-arch support (amd64, arm64)

🚀 Running Locally with Podman

This section explains how to run Red Hat Chat locally using Podman and Podman Compose.

Prerequisites

Install Podman:

# RHEL/Fedora
sudo dnf install podman
   
# Ubuntu/Debian
sudo apt-get install podman
   
# Windows
# Download Podman Desktop from https://podman-desktop.io/

Install Podman Compose:

# Using pip
pip3 install podman-compose
   
# Or use native podman compose (Podman 4.0+)
# No additional installation needed

Verify Installation:

podman --version
podman compose version  # or podman-compose --version

Quick Start

Clone the repository:

git clone https://github.com/maximilianoPizarro/LibreChat-RedHat.git
cd LibreChat-RedHat

Configure environment variables:

# Copy example environment file
cp env.example.podman .env
   
# Edit .env and configure required variables:
# - MEILI_MASTER_KEY (generate a secure key)
# - JWT_SECRET (generate a secure key)
# - JWT_REFRESH_SECRET (generate a secure key)

Start services with Podman Compose:

# Using native podman compose (recommended)
podman compose -f podman-compose.yml up -d
   
# Or using podman-compose
podman-compose -f podman-compose.yml up -d

Access the application:
- Open your browser and navigate to: http://localhost:3080
- The application should be running with all services (MongoDB, Meilisearch, etc.)

Once you access the application, you’ll see the welcome screen. Red Hat Chat supports multiple authentication methods:

Email Login: Use your email address and password
User Registration: Create a new account (if registration is enabled)
OAuth2: Login with external providers (if configured)
LDAP: Enterprise authentication (if configured)

Using Pre-built Container Image

Alternatively, you can use the pre-built image from Quay.io:

# Pull the latest image
podman pull quay.io/maximilianopizarro/redhat-chat:latest

# Run with podman-compose (recommended)
# Edit podman-compose.yml to use the image instead of building

Or run directly with Podman:

podman run -d \
  --name redhat-chat \
  -p 3080:3080 \
  -e MONGO_URI=mongodb://mongodb:27017/LibreChat \
  -e MEILI_HOST=http://meilisearch:7700 \
  quay.io/maximilianopizarro/redhat-chat:latest

Container Services

The podman-compose.yml includes the following services:

redhat-chat: Main application container
mongodb: MongoDB database for data storage
meilisearch: Search engine for conversation search
vectordb: PostgreSQL with pgvector for RAG capabilities (optional)
rag_api: RAG API service (optional, requires OPENAI_API_KEY)

Managing Services

# Start services
podman compose -f podman-compose.yml up -d

# Stop services
podman compose -f podman-compose.yml down

# View logs
podman compose -f podman-compose.yml logs -f

# Rebuild and restart
podman compose -f podman-compose.yml up -d --build

Troubleshooting

Port already in use:

# Change PORT in .env file
PORT=3081

MongoDB connection issues:

# Check MongoDB container
podman ps | grep mongodb
podman logs chat-mongodb

Podman socket connection:

# On Linux, enable podman socket
systemctl --user enable --now podman.socket
export DOCKER_HOST=unix://$XDG_RUNTIME_DIR/podman/podman.sock

☸️ Deploying on OpenShift from Catalog

This section explains how to deploy Red Hat Chat on OpenShift using the container image from Quay.io through the OpenShift catalog.

💡 Tip: The Helm chart for Red Hat Chat (LibreChat) is pre-installed in many OpenShift clusters and can be found in the OpenShift Developer Catalog under “Helm Charts”. For comprehensive Helm chart documentation, advanced configuration, and deployment strategies, visit Red Hat Chat Helm Chart Documentation.

Red Hat Chat Helm Chart available in OpenShift Developer Catalog

The Helm chart for Red Hat Chat (LibreChat) is available in the OpenShift Developer Catalog under "Helm Charts"

Prerequisites

Access to an OpenShift cluster (4.x or later)
Permissions to create deployments, services, and routes in a namespace
OpenShift CLI (oc) installed (optional, can use web console)

Method 1: Using OpenShift Web Console (Recommended)

Step 1: Access the Developer Catalog

Log in to your OpenShift web console
Navigate to Developer perspective (top menu)
Click on +Add in the left sidebar
Select Container Images from the catalog

Screenshot: OpenShift Developer Catalog - Navigate to +Add → Container Images

In the OpenShift web console, go to Developer perspective → +Add → Container Images

Step 2: Search for the Container Image

In the Image name field, enter:

quay.io/maximilianopizarro/redhat-chat:latest

Click Search or press Enter
Select the image from the results

Screenshot: Search results showing quay.io/maximilianopizarro/redhat-chat:latest

The search will display the Red Hat Chat image from Quay.io

Step 3: Configure the Deployment

Application Name: Enter redhat-chat (or your preferred name)
Name: This will be auto-filled, you can modify if needed
Resource Type: Select Deployment
Namespace: Choose or create a namespace for the deployment

Step 4: Configure Container Settings

Container Port: Set to 3080 (default application port)
Environment Variables: Add required environment variables:
- MONGO_URI: Connection string to MongoDB
- MEILI_HOST: Meilisearch host URL
- JWT_SECRET: Generate a secure JWT secret
- JWT_REFRESH_SECRET: Generate a secure refresh secret
- MEILI_MASTER_KEY: Generate a secure Meilisearch master key
Tip: You can create a Secret first and reference it:
```
oc create secret generic redhat-chat-secrets \
  --from-literal=jwt-secret=$(openssl rand -hex 32) \
  --from-literal=jwt-refresh-secret=$(openssl rand -hex 32) \
  --from-literal=meili-master-key=$(openssl rand -hex 16) \
  -n your-namespace
```
Resource Limits: Configure CPU and memory limits as needed
Storage: Add persistent volumes if needed for data storage

Step 5: Create Route

Check Create a route to the application
Configure route settings:
- Route name: redhat-chat (or your preferred name)
- Target port: 3080
- TLS: Select Edge termination (recommended)

Step 6: Deploy

Click Create to start the deployment
Wait for the deployment to complete
Access the application using the route URL provided

Screenshot: Red Hat Chat deployment status in OpenShift

Once deployed, you'll see the deployment status and route URL in the OpenShift console

After successful deployment, you can access Red Hat Chat through the OpenShift route. The application will display the same welcome screen and login interface as shown in the local deployment:

Red Hat Chat welcome screen on OpenShift

Method 2: Using OpenShift CLI (oc)

oc login --server=<your-openshift-server-url>

Step 2: Create Namespace

oc create namespace redhat-chat
oc project redhat-chat

Step 3: Create Secrets

# Generate secure keys
JWT_SECRET=$(openssl rand -hex 32)
JWT_REFRESH_SECRET=$(openssl rand -hex 32)
MEILI_MASTER_KEY=$(openssl rand -hex 16)

# Create secret
oc create secret generic redhat-chat-secrets \
  --from-literal=jwt-secret=$JWT_SECRET \
  --from-literal=jwt-refresh-secret=$JWT_REFRESH_SECRET \
  --from-literal=meili-master-key=$MEILI_MASTER_KEY

Step 4: Create Deployment

oc new-app \
  --name=redhat-chat \
  --image=quay.io/maximilianopizarro/redhat-chat:latest \
  --env=MONGO_URI=mongodb://mongodb:27017/LibreChat \
  --env=MEILI_HOST=http://meilisearch:7700 \
  --env-from=secret/redhat-chat-secrets

Step 5: Expose Service

# Create route
oc expose service redhat-chat

# Or create route with custom settings
cat <<EOF | oc apply -f -
apiVersion: route.openshift.io/v1
kind: Route
metadata:
  name: redhat-chat
  namespace: redhat-chat
spec:
  to:
    kind: Service
    name: redhat-chat
  port:
    targetPort: 3080
  tls:
    termination: edge
    insecureEdgeTerminationPolicy: Redirect
EOF

Step 6: Get Route URL

oc get route redhat-chat -o jsonpath='{.spec.host}'

Method 3: Using Helm Chart

The project includes a Helm chart for easier deployment. Note: The Helm chart for Red Hat Chat (LibreChat) is pre-installed in many OpenShift clusters and can be found in the OpenShift Developer Catalog under “Helm Charts”.

Using Pre-installed Helm Chart in OpenShift

Access the Developer Catalog:
- Log in to your OpenShift web console
- Navigate to Developer perspective
- Click on +Add → Helm Charts
- Search for “librechat” or “redhat-chat”

Red Hat Chat Helm Chart in OpenShift Developer Catalog

The Helm chart for Red Hat Chat (LibreChat) is available in the OpenShift Developer Catalog

Install from Catalog:
- Select the Helm chart from the catalog
- Configure the values (image repository, environment variables, etc.)
- Click Install to deploy

Using Helm CLI

If you prefer using the Helm CLI or the chart is not available in your cluster’s catalog:

# Add the chart repository or use local chart
helm install redhat-chat ./helm/librechat \
  --namespace redhat-chat \
  --create-namespace \
  --set image.repository=quay.io/maximilianopizarro/redhat-chat \
  --set image.tag=latest

Additional Helm Chart Documentation

For comprehensive Helm chart documentation, advanced configuration options, environment variables, OpenID Connect setup, and deployment strategies, visit:

📚 Red Hat Chat Helm Chart Documentation

The documentation includes:

Complete parameter reference
Environment variables configuration
OpenID Connect (OIDC) setup with Red Hat SSO/Keycloak
LLM endpoint configuration
Resource limits and GPU configuration
ArgoCD deployment strategies
Troubleshooting guides

Deploying Dependencies

Red Hat Chat requires MongoDB and Meilisearch. You can deploy them using:

Option 1: Using OpenShift Catalog

Deploy MongoDB from the catalog
Deploy Meilisearch from the catalog
Update environment variables in Red Hat Chat deployment to point to these services

Option 2: Using Helm

The Helm chart includes optional dependencies:

helm install redhat-chat ./helm/librechat \
  --namespace redhat-chat \
  --create-namespace \
  --set mongodb.enabled=true \
  --set meilisearch.enabled=true \
  --set image.repository=quay.io/maximilianopizarro/redhat-chat \
  --set image.tag=latest

Verifying Deployment

# Check deployment status
oc get deployment redhat-chat -n redhat-chat

# Check pods
oc get pods -n redhat-chat -l app=redhat-chat

# View logs
oc logs -f deployment/redhat-chat -n redhat-chat

# Check route
oc get route redhat-chat -n redhat-chat

Updating the Application

To update to a newer version:

# Update image tag in deployment
oc set image deployment/redhat-chat \
  redhat-chat=quay.io/maximilianopizarro/redhat-chat:latest \
  -n redhat-chat

# Or using the web console:
# 1. Go to Deployment → redhat-chat
# 2. Click Actions → Edit deployment
# 3. Update the image tag
# 4. Save

Troubleshooting

Image pull errors:

# Check if image is accessible
podman pull quay.io/maximilianopizarro/redhat-chat:latest

# Check image pull secrets
oc get secrets -n redhat-chat | grep registry

Deployment not starting:

# Check deployment events
oc describe deployment redhat-chat -n redhat-chat

# Check pod events
oc describe pod <pod-name> -n redhat-chat

# Check logs
oc logs <pod-name> -n redhat-chat

Route not accessible:

# Check route status
oc get route redhat-chat -n redhat-chat -o yaml

# Test service directly
oc get svc redhat-chat -n redhat-chat
oc port-forward svc/redhat-chat 3080:3080 -n redhat-chat

🎨 Red Hat Design System Integration

This application uses the Red Hat Design System (RHDS) for consistent UI components and styling. The frontend is built with:

Red Hat Design System Elements (@rhds/elements@4.0.0) - Web components following Red Hat design standards
Red Hat Design Tokens - CSS variables for colors, spacing, typography, and more
Red Hat Icons - Consistent iconography across the application

The design system is loaded via CDN (jsDelivr) for development, and can be configured to use the Red Hat CDN for production deployments on *.redhat.com domains.

📦 Container Image Details

Image Repository

Registry: Quay.io
Repository: maximilianopizarro/redhat-chat
Full Image Path: quay.io/maximilianopizarro/redhat-chat:latest

Available Tags

Check the Quay.io repository for all available tags:

latest - Latest stable release
v0.8.2-rc1 - Specific version tag
Other version tags as they become available

Image Architecture

Base Image: registry.access.redhat.com/ubi8/nodejs-20:latest
Architecture: Multi-arch (amd64, arm64)
Size: ~500MB (compressed)

Pulling the Image

# Pull latest
podman pull quay.io/maximilianopizarro/redhat-chat:latest

# Pull specific version
podman pull quay.io/maximilianopizarro/redhat-chat:v0.8.2-rc1

# Verify image
podman images | grep redhat-chat

🔧 Configuration

librechat.yaml Configuration

Red Hat Chat uses a librechat.yaml configuration file to customize AI endpoints, file handling, rate limits, and more. Below are examples for configuring Llama Stack endpoints.

Basic Configuration Structure

The librechat.yaml file supports multiple configuration sections:

File Configuration: Limits for file uploads per endpoint
Rate Limits: IP and user-based rate limiting
Endpoints: Custom AI model endpoints (Ollama, Llama models, etc.)
Assistants: Configuration for AI assistants

Example 1: Local Ollama with Llama Stack

This example configures Red Hat Chat to use a local Ollama instance running Llama models:

version: 1.0.3
cache: true

fileConfig:
  endpoints:
    assistants:
      fileLimit: 5
      fileSizeLimit: 10  # Maximum size for an individual file in MB
      totalSizeLimit: 50  # Maximum total size for all files in a single request in MB
      supportedMimeTypes:
        - "image/.*"
        - "application/pdf"
    openAI:
      disabled: true  # Disables file uploading to the OpenAI endpoint
    default:
      totalSizeLimit: 20
  serverFileSizeLimit: 100  # Global server file size limit in MB
  avatarSizeLimit: 2  # Limit for user avatar image size in MB

rateLimits:
  fileUploads:
    ipMax: 100
    ipWindowInMinutes: 60  # Rate limit window for file uploads per IP
    userMax: 50
    userWindowInMinutes: 60  # Rate limit window for file uploads per user

endpoints:
  assistants:
    disableBuilder: false  # Disable Assistants Builder Interface by setting to `true`
    pollIntervalMs: 750  # Polling interval for checking assistant updates
    timeoutMs: 180000  # Timeout for assistant operations
    fileLimit: 5
    fileSizeLimit: 10   # Maximum size for an individual file in MB
    totalSizeLimit: 50  # Maximum total size for all files in a single request in MB
    supportedMimeTypes:
      - "image/.*"
      - "application/pdf"
  custom:
    # Local Ollama instance with Llama2
    - name: "Ollama"
      apiKey: "IiI="  # Base64 encoded empty string (Ollama doesn't require auth)
      baseURL: "http://librechat-ollama:11434/v1"  # Adjust if Ollama is on a different host
      models:
        default: ["llama2", "llama3", "llama3.2:3b-instruct"]  # Available models
        fetch: true  # Fetch available models from Ollama API
      titleConvo: true
      titleModel: "llama2"
      summarize: false
      summaryModel: "llama2"

Example 2: Remote Llama Model via Red Hat AI Services

This example shows how to connect to a remote Llama model hosted on Red Hat AI Services:

version: 1.0.3
cache: true

fileConfig:
  endpoints:
    assistants:
      fileLimit: 5
      fileSizeLimit: 10
      totalSizeLimit: 50
      supportedMimeTypes:
        - "image/.*"
        - "application/pdf"
    openAI:
      disabled: true
    default:
      totalSizeLimit: 20
  serverFileSizeLimit: 100
  avatarSizeLimit: 2

rateLimits:
  fileUploads:
    ipMax: 100
    ipWindowInMinutes: 60
    userMax: 50
    userWindowInMinutes: 60

endpoints:
  assistants:
    disableBuilder: false
    pollIntervalMs: 750
    timeoutMs: 180000
    fileLimit: 5
    fileSizeLimit: 10
    totalSizeLimit: 50
    supportedMimeTypes:
      - "image/.*"
      - "application/pdf"
  custom:
    # Remote Llama 3.2 3B model via Red Hat AI Services
    - name: "llama-3-2-3b"
      apiKey: "your-api-key-here"  # Replace with your actual API key
      baseURL: "https://llama-3-2-3b-maas-apicast-production.apps.prod.rhoai.rh-aiservices-bu.com:443/v1"
      models:
        default: ["llama-3-2-3b"]
        fetch: false  # Don't fetch, use explicit model list
      titleConvo: true
      titleModel: "llama-3-2-3b"
      summarize: false
      summaryModel: "llama-3-2-3b"

Example 3: Multiple Llama Models (Local + Remote)

This example demonstrates configuring both local Ollama and remote Llama models:

version: 1.0.3
cache: true

fileConfig:
  endpoints:
    assistants:
      fileLimit: 5
      fileSizeLimit: 10
      totalSizeLimit: 50
      supportedMimeTypes:
        - "image/.*"
        - "application/pdf"
    openAI:
      disabled: true
    default:
      totalSizeLimit: 20
  serverFileSizeLimit: 100
  avatarSizeLimit: 2

rateLimits:
  fileUploads:
    ipMax: 100
    ipWindowInMinutes: 60
    userMax: 50
    userWindowInMinutes: 60

endpoints:
  assistants:
    disableBuilder: false
    pollIntervalMs: 750
    timeoutMs: 180000
    fileLimit: 5
    fileSizeLimit: 10
    totalSizeLimit: 50
    supportedMimeTypes:
      - "image/.*"
      - "application/pdf"
  custom:
    # Local Ollama - Lightweight models for development
    - name: "Ollama Local"
      apiKey: "IiI="
      baseURL: "http://localhost:11434/v1"  # Local Ollama instance
      models:
        default: ["llama3.2:3b-instruct", "llama3.2:1b-instruct"]  # Lightweight models
        fetch: true
      titleConvo: true
      titleModel: "llama3.2:3b-instruct"
      summarize: false
      summaryModel: "llama3.2:3b-instruct"
    
    # Remote Llama - Production models
    - name: "Llama 3.2 3B Production"
      apiKey: "${LLAMA_API_KEY}"  # Use environment variable
      baseURL: "https://llama-3-2-3b-maas-apicast-production.apps.prod.rhoai.rh-aiservices-bu.com:443/v1"
      models:
        default: ["llama-3-2-3b"]
      titleConvo: true
      titleModel: "llama-3-2-3b"
      summarize: false
      summaryModel: "llama-3-2-3b"
    
    # Another remote model option
    - name: "Llama 3 70B"
      apiKey: "${LLAMA_70B_API_KEY}"
      baseURL: "https://your-llama-70b-endpoint.com/v1"
      models:
        default: ["llama-3-70b-instruct"]
      titleConvo: true
      titleModel: "llama-3-70b-instruct"
      summarize: true
      summaryModel: "llama-3-70b-instruct"

Configuration Notes

For Podman/Container Deployments:

Mount the configuration file: Ensure librechat.yaml is mounted as a volume in your container: ```yaml volumes:
- type: bind source: ./librechat.yaml target: /app/librechat.yaml read_only: true ```
Environment Variables: You can use environment variables in the YAML file:
```
apiKey: "${LLAMA_API_KEY}"
```
Then set the variable in your .env file or container environment.

Network Configuration: For local Ollama, use the container service name:

baseURL: "http://librechat-ollama:11434/v1"  # Service name in podman-compose

For OpenShift Deployments:

ConfigMap: Create a ConfigMap from your librechat.yaml:

oc create configmap librechat-config --from-file=librechat.yaml -n redhat-chat

Mount ConfigMap: Mount it in your deployment: ```yaml volumeMounts:
- name: config mountPath: /app/librechat.yaml subPath: librechat.yaml volumes:
- name: config configMap: name: librechat-config ```

Secrets for API Keys: Store API keys in Secrets:

oc create secret generic llama-api-keys \
  --from-literal=llama-api-key=your-key-here \
  -n redhat-chat

Available Llama Models

Common Llama models you can use with Red Hat Chat:

llama2 - Meta’s Llama 2 model (7B, 13B, 70B variants)
llama3 - Meta’s Llama 3 model (8B, 70B variants)
llama3.2:1b-instruct - Lightweight 1B parameter model
llama3.2:3b-instruct - 3B parameter model (good balance)
llama-3-2-3b - Red Hat AI Services hosted model
llama-3-70b-instruct - Large 70B parameter model

Troubleshooting Configuration

Model not appearing in UI:

Check that fetch: true is set or models are listed in default
Verify the baseURL is correct and accessible
Check container logs for connection errors

API Key issues:

Ensure API keys are correctly set in environment variables or secrets
For Ollama local, use apiKey: "IiI=" (base64 empty string)

Connection errors:

Verify network connectivity between containers
Check firewall rules for external endpoints
Ensure service names match in podman-compose.yml

Environment Variables

Key environment variables for Red Hat Chat:

Required:

MONGO_URI - MongoDB connection string
MEILI_HOST - Meilisearch host URL
JWT_SECRET - JWT signing secret (min 32 chars)
JWT_REFRESH_SECRET - JWT refresh secret (min 32 chars)
MEILI_MASTER_KEY - Meilisearch master key (min 16 chars)

Optional:

PORT - Application port (default: 3080)
DOMAIN - Application domain
ALLOW_REGISTRATION - Enable user registration (default: true)
ALLOW_EMAIL_LOGIN - Enable email login (default: true)

See env.example.podman for a complete list of configuration options.

Generating Secure Keys

# Generate JWT secrets (32+ characters)
openssl rand -hex 32

# Generate Meilisearch master key (16+ characters)
openssl rand -hex 16

📚 Additional Resources

Red Hat Resources:

Red Hat Design System: ux.redhat.com
Red Hat UBI: developers.redhat.com/products/rhel/ubi
Podman: podman.io

Original LibreChat Resources:

GitHub Repo: github.com/danny-avila/LibreChat
Website: librechat.ai
Documentation: librechat.ai/docs

Container Registry:

Quay.io Repository: quay.io/repository/maximilianopizarro/redhat-chat

🆘 Support

For issues, questions, or contributions:

GitHub Issues: github.com/maximilianoPizarro/LibreChat-RedHat/issues
Documentation: See project README.md for detailed setup instructions

📝 License

This project is open source. Please refer to the LICENSE file in the repository for details.

Red Hat Chat - Powered by LibreChat with Red Hat Design System

Built with ❤️ using Red Hat certified container images

📋 TL;DR

📖 Overview

Key Features

Container Image

🚀 Running Locally with Podman

Prerequisites

Quick Start

First Steps: Login and Registration

Using Pre-built Container Image

Container Services

Managing Services

Troubleshooting

☸️ Deploying on OpenShift from Catalog

Prerequisites

Method 1: Using OpenShift Web Console (Recommended)

Step 1: Access the Developer Catalog

Step 2: Search for the Container Image

Step 3: Configure the Deployment

Step 4: Configure Container Settings

Step 5: Create Route

Step 6: Deploy

Method 2: Using OpenShift CLI (oc)

Step 1: Login to OpenShift

Step 2: Create Namespace

Step 3: Create Secrets

Step 4: Create Deployment

Step 5: Expose Service

Step 6: Get Route URL

Method 3: Using Helm Chart

Using Pre-installed Helm Chart in OpenShift

Using Helm CLI

Additional Helm Chart Documentation

Deploying Dependencies

Option 1: Using OpenShift Catalog

Option 2: Using Helm

Verifying Deployment

Updating the Application

Troubleshooting

🎨 Red Hat Design System Integration

📦 Container Image Details

Image Repository

Available Tags

Image Architecture

Pulling the Image

🔧 Configuration

librechat.yaml Configuration

Basic Configuration Structure

Example 1: Local Ollama with Llama Stack

Example 2: Remote Llama Model via Red Hat AI Services

Example 3: Multiple Llama Models (Local + Remote)

Configuration Notes

Available Llama Models

Troubleshooting Configuration

Environment Variables

Generating Secure Keys

📚 Additional Resources

🆘 Support

📝 License