codinit-dev · Gerome-Elassaad · Jan 22, 2026 · Jan 22, 2026 · Jan 22, 2026 · Jan 22, 2026
diff --git a/essentials/customization.mdx b/essentials/customization.mdx
@@ -65,13 +65,13 @@
 Choose from 9 professionally selected fonts:

 - **Inter** - Default, modern sans-serif
 - **Roboto** - Google's popular sans-serif
 - **Open Sans** - Friendly and readable
 - **Montserrat** - Geometric sans-serif
 - **Poppins** - Geometric with playful character
 - **Lato** - Humanist sans-serif
 - **JetBrains Mono** - Monospace for code
 - **Raleway** - Elegant display font
 - **Lora** - Serif for traditional look

 <Tip>
@@ -148,18 +148,18 @@

 ### preset themes

 11 preset design themes to jumpstart customization:

 - **Minimal** - Clean, distraction-free design
 - **Modern** - Contemporary with bold accents
 - **Carbon** - IBM Carbon-inspired system
 - **Material** - Google Material Design principles
 - **Flat** - Flat design with vibrant colors
 - **Neobrutalism** - Bold borders and stark contrasts
 - **Glassmorphism** - Frosted glass effects
 - **Claymorphism** - Soft, 3D clay-like elements
 - **Retro** - Nostalgic design patterns
 - **Neumorphism** - Soft UI with subtle shadows
 - **Cyberpunk** - Futuristic neon aesthetics

 ## control panel settings
@@ -227,20 +227,28 @@

 Configure local AI inference:

 - **Ollama**: Run models locally via Ollama
 - **LM Studio**: Desktop application for local models
 - **OpenAI-compatible**: Custom OpenAI-compatible endpoints
 
 **Local provider settings:**
-- Custom base URL
-- Connection testing
-- Model discovery
+- Automatic detection of running providers
+- Custom base URL configuration
+- Real-time connection status
+- Model discovery and listing
 - Enable/disable toggle
 
 <Tip>
-Local providers require the respective software running on your machine. Connection status is shown in real-time.
+CodinIT automatically detects when Ollama (port 11434) or LM Studio (port 1234) are running on your machine. Detected providers are automatically enabled and appear in your model selector. Connection status updates in real-time.
 </Tip>
 
+**Auto-detection features:**
+- Scans for local providers on startup
+- Automatically enables detected providers
+- Disables providers when they go offline
+- Shows loading states during detection
+- Integrates seamlessly with model selection
+
 ### API keys
 
 Centralized API key management for all AI providers:
@@ -264,17 +272,17 @@
 - Clone templates
 - Push local projects

 **vercel**
 - Connect Vercel account
 - Deploy projects
 - View deployment status

 **netlify**
 - Connect Netlify account
 - Deploy sites
 - Manage domains

 **cloudflare**
 - Configure Cloudflare Pages
 - Deploy applications

@@ -387,7 +395,7 @@

 ### theme persistence

 Theme preference is stored in browser localStorage and persists across sessions.

 ## keyboard shortcuts

@@ -404,7 +412,7 @@

 ## browser storage

 All CodinIT customization is stored in browser localStorage:

 **Stored data:**
 - Design palette preferences
@@ -439,7 +447,7 @@

 ### settings not saving

 - Check browser localStorage permissions
 - Disable browser extensions that block storage
 - Try a different browser
 - Clear cache and reload

diff --git a/providers/ollama.mdx b/providers/ollama.mdx
@@ -1,16 +1,16 @@
 ---
 title: "Ollama"
 description: "Run AI models locally with Ollama for privacy and offline access."
 ---

 Run models locally using Ollama for privacy, offline access, and control. Requires initial setup and sufficient hardware.

 **Website:** [https://ollama.com/](https://ollama.com/)

 ## Setup

 1. **Install Ollama:** Download from [ollama.com](https://ollama.com/) and install
 2. **Start Ollama:** Run `ollama serve` in terminal
 3. **Download a model:**
   ```bash
   ollama pull qwen2.5-coder:32b
@@ -25,7 +25,7 @@
 ## Configuration in CodinIT

 1. Click the settings icon (⚙️) in CodinIT
 2. Select "ollama" as the API Provider
 3. Enter your saved model name
 4. (Optional) Set base URL if not using default `http://localhost:11434`

@@ -38,9 +38,25 @@
 
 See [Ollama model library](https://ollama.com/library) for full list.
 
+## Dynamic Context Windows
+
+CodinIT automatically calculates optimal context windows based on model parameter size:
+
+- **70B+ models:** 32k context window (e.g., Llama 70B)
+- **30B+ models:** 16k context window
+- **7B+ models:** 8k context window
+- **Smaller models:** 4k context window (default)
+
+**Special model families:**
+- Llama 70B models: 32k context
+- Llama 405B models: 128k context
+
+Model labels in CodinIT show both parameter size and context window (e.g., "qwen2.5-coder:32b (32B, 16k ctx)").
+
 ## Notes
 
-- **Context window:** Minimum 12,000 tokens recommended, 32,000 ideal
+- **Auto-detection:** CodinIT automatically detects Ollama running on port 11434
+- **Context window:** Dynamically calculated based on model capabilities
 - **Resource demands:** Large models require significant system resources
 - **Offline capability:** Works without internet after model download
 - **Performance:** May be slow on average hardware
diff --git a/providers/openrouter.mdx b/providers/openrouter.mdx
@@ -22,7 +22,15 @@ OpenRouter provides access to models from multiple providers through a single AP
 
 ## Supported Models
 
-CodinIT automatically fetches available models. See [OpenRouter Models](https://openrouter.ai/models) for the complete list.
+CodinIT automatically fetches available models. Featured models include:
+
+- **Claude Opus 4.5:** 200k context, maximum intelligence
+- **Claude Sonnet 4.5:** 1M context, highest intelligence
+- **GPT-5.2 Pro:** 400k context, latest GPT model
+- **GPT-4o:** 128k context, reliable fallback
+- **DeepSeek R1 (Free):** 163k context, free tier available
+
+See [OpenRouter Models](https://openrouter.ai/models) for the complete list.
 
 ## Features