feat: Add comprehensive OpenAI base URL configuration support

archon-ui-main/src/components/settings/RAGSettings.tsx

@@ -17,6 +17,7 @@ interface RAGSettingsProps {
     USE_RERANKING: boolean;
     LLM_PROVIDER?: string;
     LLM_BASE_URL?: string;
+    OPENAI_BASE_URL?: string;
     EMBEDDING_MODEL?: string;
     // Crawling Performance Settings
     CRAWL_BATCH_SIZE?: number;
@@ -85,6 +86,20 @@ export const RAGSettings = ({
               />
             </div>
           )}
+          {ragSettings.LLM_PROVIDER === 'openai' && (
+            <div>
+              <Input
+                label="OpenAI Base URL (optional)"
+                value={ragSettings.OPENAI_BASE_URL || ''}
+                onChange={e => setRagSettings({
+                  ...ragSettings,
+                  OPENAI_BASE_URL: e.target.value
+                })}
+                placeholder="https://api.openai.com/v1"
+                accentColor="green"
+              />
+            </div>
+          )}
           <div className="flex items-end">
             <Button 
               variant="outline" 

docs/docs/api-reference.mdx

@@ -1466,6 +1466,7 @@ Retrieve all RAG configuration settings including LLM provider options.
     "USE_RERANKING": "true",
     "LLM_PROVIDER": "openai",
     "LLM_BASE_URL": null,
+    "OPENAI_BASE_URL": "http://localhost:8000/v1",
     "EMBEDDING_MODEL": null
   }
 }
@@ -1475,11 +1476,11 @@ Retrieve all RAG configuration settings including LLM provider options.
 
 | Setting | Type | Description |
 |---------|------|-------------|
-| `LLM_PROVIDER` | string | Provider choice: `openai`, `ollama`, `google` |
-| `LLM_BASE_URL` | string | Custom base URL (required for Ollama) |
-| `EMBEDDING_MODEL` | string | Override default embedding model |
-| `MODEL_CHOICE` | string | Chat model for summaries and contextual embeddings |
-
+| `LLM_PROVIDER`     | string | Provider choice: `openai`, `ollama`, `google` |
+| `LLM_BASE_URL`     | string | Custom base URL (required for Ollama) |
+| `OPENAI_BASE_URL`  | string | Custom OpenAI endpoint for proxies (LiteLLM, Azure OpenAI) — applies system-wide |
+| `EMBEDDING_MODEL`  | string | Override default embedding model |
+| `MODEL_CHOICE`     | string | Chat model for summaries and contextual embeddings |
 ## 💬 Agent Chat API
 
 <Admonition type="warning" title="Integration Note">

docs/docs/configuration.mdx

@@ -141,10 +141,57 @@ Configure in **Settings → RAG Settings → LLM Provider**
 
 <Admonition type="tip" title="Provider Setup">
 - **API Keys**: Add provider-specific keys in Settings → API Keys
+- **Custom Base URLs**: Configure custom endpoints for OpenAI-compatible proxies
 - **Ollama**: Requires local installation and custom base URL
 - **Models**: Each provider has different model naming conventions
 </Admonition>
 
+#### OpenAI Base URL Configuration
+
+For advanced use cases, you can configure a custom OpenAI base URL to route **all OpenAI API requests** through:
+- **LiteLLM proxy** for multi-provider access
+- **Azure OpenAI** endpoints
+- **Corporate proxies** or custom gateways
+- **Local OpenAI-compatible servers**
+
+<Admonition type="info" title="Centralized Configuration">
+The `OPENAI_BASE_URL` setting applies consistently across **all services**:
+- PydanticAI agents (RAG, document processing, task management)
+- LLM provider service (embeddings, contextual processing)
+- Background tasks and batch operations
+
+This ensures consistent proxy routing for all OpenAI API usage throughout the system.
+</Admonition>
+
+<Tabs>
+<TabItem value="ui" label="UI Configuration" default>
+
+1. Navigate to **Settings → RAG Settings**
+2. Select **OpenAI** as your LLM Provider
+3. Enter your custom **OpenAI Base URL** (optional)
+4. Examples:
+   - LiteLLM: `http://localhost:8000/v1`
+   - Azure OpenAI: `https://your-resource.openai.azure.com/openai/deployments/your-deployment`
+   - Custom proxy: `https://api.yourcompany.com/openai/v1`
+
+</TabItem>
+<TabItem value="env" label="Environment Variables">
+
+```bash
+# Optional - Custom OpenAI endpoint
+OPENAI_BASE_URL=http://localhost:8000/v1
+
+# Required when using custom base URL
+OPENAI_API_KEY=your-api-key-here
+```
+
+<Admonition type="warning" title="Security Requirements">
+When using a custom `OPENAI_BASE_URL`, an API key must be provided for security reasons. This prevents accidental traffic leaks to public endpoints when a proxy is explicitly configured.
+</Admonition>
+
+</TabItem>
+</Tabs>
+
 ### Socket.IO Configuration
 
 Archon uses Socket.IO for all real-time features. The default configuration works out of the box:

docs/docs/rag.mdx

@@ -303,10 +303,66 @@ USE_RERANKING=true
 LLM_PROVIDER=openai
 MODEL_CHOICE=gpt-4o-mini
 EMBEDDING_MODEL=text-embedding-3-small
-# Pros: Best accuracy, reliable API
+
+# Optional: Custom OpenAI endpoint
+OPENAI_BASE_URL=https://api.openai.com/v1
+
+# Pros: Best accuracy, reliable API, proxy support
 # Cons: Cost per query
 ```
 
+#### OpenAI Proxy Configuration
+
+Configure custom endpoints for advanced routing scenarios. These settings apply to **all OpenAI API usage** including RAG queries, embeddings, and background processing:
+
+<Tabs>
+<TabItem value="litellm" label="LiteLLM Proxy" default>
+
+```bash
+# LiteLLM proxy for multi-provider access
+LLM_PROVIDER=openai
+OPENAI_BASE_URL=http://localhost:8000/v1
+MODEL_CHOICE=gpt-4o-mini  # Route through LiteLLM
+OPENAI_API_KEY=your-litellm-key
+
+# Benefits:
+# - Multi-provider routing (OpenAI, Anthropic, etc.)
+# - Cost tracking and rate limiting
+# - Centralized API management
+```
+
+**LiteLLM Setup:**
+```bash
+# Install LiteLLM
+pip install litellm[proxy]
+
+# Start proxy server
+litellm --port 8000 --config litellm_config.yaml
+```
+
+</TabItem>
+<TabItem value="azure" label="Azure OpenAI">
+
+
+</TabItem>
+<TabItem value="corporate" label="Corporate Proxy">
+
+```bash
+# Corporate gateway/proxy
+LLM_PROVIDER=openai
+OPENAI_BASE_URL=https://api.yourcompany.com/openai/v1
+MODEL_CHOICE=gpt-4o-mini
+OPENAI_API_KEY=your-internal-key
+
+# Benefits:
+# - Centralized billing and monitoring
+# - Security and compliance enforcement
+# - Custom authentication and authorization
+```
+
+</TabItem>
+</Tabs>
+
 ### Google Gemini
 ```bash
 LLM_PROVIDER=google
@@ -368,6 +424,13 @@ EMBEDDING_MODEL=nomic-embed-text
 3. Use caching more aggressively
 4. Consider switching to local models (Ollama)
 
+### API Key Management
+- **Separate Keys**: Use different API keys for proxy vs direct OpenAI access
+- **Key Rotation**: Regularly rotate API keys, especially for proxy endpoints
+- **Environment Variables**: Never hardcode keys in configuration files
+- **Least Privilege**: Proxy keys should have minimal required permissions
+- **No Secrets in Logs**: Disable logging of Authorization headers and redact API keys in all logs.
+
 ## 🎯 Best Practices
 
 ### Content Optimization