Arka MetaBrain
Intelligent AI routing system that automatically selects the optimal model, applies enhanced context, and suggests actions for every task.
Premium Intelligence Layer
MetaBrain provides intelligent routing + context enhancement + action suggestions. Note: Uses 10× token multiplier due to advanced processing.
Enable MetaBrainWhat is MetaBrain?
MetaBrain is ARKA AI's intelligent orchestration layer that sits above individual AI models. Instead of manually selecting GPT-4, Claude, or other models, MetaBrain analyzes your request and automatically routes it to the most capable model for that specific task.
Beyond routing, MetaBrain enhances every interaction with additional context, suggests follow-up actions, and ensures optimal performance.
Smart Routing
Automatically selects the best AI model based on task type, complexity, and required capabilities.
Context Enhancement
Enriches requests with data from ArkaMind, Live Context, and integrations automatically.
Action Suggestions
Proactively suggests next steps, automations, and relevant tools based on your goals.
How MetaBrain Works
1. Task Analysis
MetaBrain analyzes your request to determine task type (creative, analytical, coding, etc.) and required capabilities (vision, reasoning, speed).
2. Intelligent Routing
Selects the optimal model: GPT-4o for speed, GPT-4 for complex reasoning, Claude for long documents, etc.
3. Context Gathering
Automatically retrieves relevant memories from ArkaMind, live business data, and past conversations to enhance context.
4. Enhanced Processing
Sends enriched request to selected model with optimized prompt engineering for best results.
5. Action Synthesis
Analyzes response and suggests relevant follow-up actions, automations, or integrations.
Routing Logic
MetaBrain uses sophisticated heuristics to select the best model:
| Task Type | Selected Model | Reason |
|---|---|---|
| Quick questions | GPT-4o-mini | Fastest, cost-effective |
| Complex reasoning | GPT-4 | Best logic capabilities |
| Long documents | Claude 3 Opus | 200K token context |
| Image analysis | GPT-4o | Advanced vision |
| Creative writing | Claude 3.5 Sonnet | Natural, nuanced prose |
| Code generation | GPT-4 | Best coding performance |
Context Enhancement
MetaBrain automatically enriches your requests with relevant context:
🧠 ArkaMind Integration
Pulls relevant memories and past decisions automatically without you having to mention them.
📊 Live Context Data
Includes real-time business metrics from integrated systems when relevant to your query.
📦 File References
Automatically includes content from ArkaStorage files when you reference documents by name.
Action Suggestions
MetaBrain proactively suggests next steps after processing your request:
🤖Automation Suggestions
"Would you like to create a workflow to automate this task?" with one-click setup.
🔌Integration Recommendations
"Connect Stripe to get real-time revenue data for future queries" when financial questions are asked.
📊Analysis Prompts
"Would you like me to analyze trends in this data?" or "Should I create a chart from these numbers?"
Token Usage & Pricing
⚠️ 10× Token Multiplier
MetaBrain uses 10× the normal token count due to:
- • Context gathering from multiple sources
- • Enhanced prompt engineering
- • Multi-model coordination
- • Action synthesis and suggestions
Example: A simple query using 500 tokens will count as 5,000 tokens with MetaBrain enabled.
When to Use MetaBrain
✅ Great For
- • Complex, multi-step tasks requiring optimal model selection
- • Queries needing context from multiple sources
- • Strategic decisions where quality matters more than cost
- • Learning mode - let AI teach you best practices
❌ Avoid For
- • Simple, repetitive queries
- • High-volume automated workflows
- • Budget-constrained use cases
- • When you already know which model you need
Enabling MetaBrain
- Click the model selector dropdown in your dashboard
- Toggle Arka MetaBrain at the bottom
- Confirm you understand the 10× token multiplier
- MetaBrain is now active for all new conversations
💡 Tip: Per-Conversation Toggle
You can enable/disable MetaBrain for specific conversations. Turn it on for complex tasks, off for simple queries to manage token usage.
MetaBrain in Action
Without MetaBrain:
User: "What's our revenue this month?"
AI: "I don't have access to real-time revenue data. Please check your billing dashboard."
With MetaBrain:
User: "What's our revenue this month?"
AI: "Your revenue for October 2025 is $42,350 (via Stripe Live Context). This is 18% higher than September. Would you like me to:
• Create a revenue forecast for Q4
• Analyze which plans drove the increase
• Set up an alert for daily revenue thresholds"
Best Practices
🎯 Use Selectively
Enable MetaBrain for important tasks. Disable for routine queries to manage token costs.
📊 Monitor Usage
Check your usage dashboard regularly. MetaBrain can consume tokens quickly on complex tasks.
🔄 Review Suggestions
Take advantage of action suggestions. MetaBrain learns your workflow patterns over time.