Help & Support

Comprehensive guides for using Insurgent Ink

Custom APIs

Set up custom API endpoints for specialised translation services and self-hosted AI models.

Custom API Overview

Insurgent Ink supports custom API endpoints that follow the OpenAI API format. This allows you to integrate:

🏠 Self-Hosted Models

  • • Local AI inference servers
  • • Private model deployments
  • • Enterprise AI solutions
  • • Custom-trained models

🔧 Alternative Services

  • • OpenAI-compatible APIs
  • • Specialised translation services
  • • Regional AI providers
  • • Research platforms

API Requirements

Your custom API must be compatible with the OpenAI chat completions format to work with Insurgent Ink.

✅ Required Features

  • • OpenAI-compatible API format
  • • HTTPS endpoint (secure connection)
  • • POST /chat/completions endpoint
  • • JSON request/response format
  • • Authentication support (API key or headers)

🔧 API Format Example

POST /chat/completions
{
  "model": "your-model-name",
  "messages": [
    {
      "role": "user",
      "content": "Translate this text to Spanish: Hello world"
    }
  ]
}

Setting Up Custom API

1

Prepare Your API Endpoint

Ensure your API is running and accessible via HTTPS with the correct endpoint format.

Common Endpoint Patterns:

  • • https://your-api.com/v1 (OpenAI format)
  • • https://localhost:8080/v1 (Local development)
  • • https://api.yourcompany.com/chat (Custom format)
2

Add Custom API in Insurgent Ink

Go to Settings → API Keys and select "Custom API" as the provider.

3

Configure Custom Headers (Optional)

Add any additional headers required by your API endpoint.

Format: Valid JSON object with header name-value pairs

4

Test & Save

Use the "Test Key" button to verify your API configuration before saving.

API Test Successful

Your custom API is configured correctly and ready to use.

Common Integrations

Ollama (Local AI)

Configuration:

URL: http://localhost:11434/v1
Model: llama2, codellama, etc.
API Key: ollama (or any value)

Perfect for running AI models locally with privacy and control.

Text Generation WebUI

Configuration:

URL: http://localhost:5000/v1
Model: loaded-model-name
API Key: any-value

Popular web interface for running various open-source LLMs.

LM Studio

Configuration:

URL: http://localhost:1234/v1
Model: current-loaded-model
API Key: lm-studio

User-friendly application for running LLMs locally with GUI.

Custom Enterprise API

Configuration:

URL: https://ai.company.com/v1
Headers: Custom auth tokens
Model: enterprise-model

Connect to your organisation's private AI infrastructure.

Troubleshooting Custom APIs

🔧 Connection Issues

  • • Verify the URL is accessible from your network
  • • Check if HTTPS is required (recommended)
  • • Ensure the API is running and responding
  • • Test with curl or Postman first

🔑 Authentication Errors

  • • Verify API key format and validity
  • • Check custom headers are correctly formatted JSON
  • • Ensure authentication method matches API requirements
  • • Test authentication independently

📋 Format Compatibility

  • • Confirm API follows OpenAI chat completions format
  • • Check model name is recognised by the API
  • • Verify JSON request/response structure
  • • Test with simple translation requests first

Advanced Custom API Features

Insurgent Ink's Phase 4 enhancements include powerful features for custom API integration:

🎯 Model-Specific Configuration

Per-Model Settings:
  • • Different timeout values per model
  • • Model-specific parameter adjustments
  • • Custom prompt templates
  • • Quality vs speed profiles
Dynamic Model Selection:
  • • Switch models based on content type
  • • Fallback model configuration
  • • Load balancing across models
  • • A/B testing capabilities

⚙️ Enhanced Header Management

Dynamic Headers Example:

{
  "Authorization": "Bearer {{api_key}}",
  "X-Model-Type": "translation",
  "X-Quality-Level": "high",
  "User-Agent": "Insurgent-Ink/{{version}}",
  "Accept": "application/json",
  "Content-Type": "application/json"
}

Support for variable substitution, authentication patterns, and provider-specific headers.

🔄 Intelligent Rate Limiting

Automatic Detection:
  • • Recognizes custom API rate limits
  • • Adapts to response headers
  • • Learns from error patterns
  • • Optimizes retry timing
Custom Rate Limiting:
  • • Configure specific limits per endpoint
  • • Burst allowance settings
  • • Time window customization
  • • Priority queue management

Enterprise Custom API Integration

Advanced features for enterprise deployments and large-scale custom API usage:

🏢 Multi-Tenant Support

  • • Team-specific custom API configurations
  • • Role-based access to custom endpoints
  • • Cost tracking per business unit
  • • Audit logging for compliance
  • • Centralized configuration management

📊 Monitoring & Analytics

  • • Real-time performance monitoring
  • • Custom API usage statistics
  • • Error rate tracking and alerting
  • • Cost analysis and optimization
  • • SLA compliance reporting

🔧 DevOps Integration

  • • Infrastructure as Code support
  • • CI/CD pipeline integration
  • • Environment-specific configurations
  • • Automated testing and validation
  • • Blue-green deployment support

🛡️ Enhanced Security

  • • Mutual TLS authentication
  • • IP allowlisting and restrictions
  • • Certificate pinning
  • • Encrypted credential storage
  • • Zero-trust network integration

Popular Custom API Use Cases

Real-world examples of how organizations use custom APIs with Insurgent Ink:

🏥 Healthcare Translation

Medical organizations using specialized healthcare AI models:

  • • HIPAA-compliant local deployments
  • • Medical terminology preservation
  • • Multi-language patient documentation
  • • Regulatory compliance features

⚖️ Legal Document Processing

Law firms with custom legal AI models:

  • • Jurisdiction-specific legal terminology
  • • Contract translation with precedent awareness
  • • Client confidentiality protection
  • • Regulatory compliance tracking

🏭 Manufacturing Documentation

Industrial companies with technical AI models:

  • • Technical specification translation
  • • Safety manual localization
  • • Industry-standard terminology
  • • Quality control documentation

🎮 Game Localization

Gaming companies with entertainment-focused models:

  • • Cultural adaptation and localization
  • • Character voice and tone preservation
  • • Regional content compliance
  • • Rapid iteration and updates

Security & Compliance

🔒 Security Best Practices

  • • Always use HTTPS for production APIs
  • • Implement proper certificate validation
  • • Use strong authentication tokens
  • • Monitor API usage and access logs
  • • Regular security audits and penetration testing
  • • Encrypted data transmission and storage

⚠️ Important Considerations

  • • Custom APIs may have different rate limits
  • • You control data privacy and retention
  • • Compliance responsibility varies by deployment
  • • Test thoroughly in staging environments
  • • Document API dependencies and requirements
  • • Plan for disaster recovery and failover

📋 Compliance Features

  • • GDPR data processing controls
  • • SOC 2 Type II audit support
  • • HIPAA compliance capabilities
  • • ISO 27001 security framework alignment
  • • Industry-specific regulatory support

🔄 Reliability & Monitoring

  • • Health check endpoints and monitoring
  • • Automatic failover and retry logic
  • • Performance metrics and alerting
  • • Load balancing and scaling support
  • • Circuit breaker pattern implementation