> ## Documentation Index
> Fetch the complete documentation index at: https://docs.anannas.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# FAQs

> Common questions about Anannas - The unified AI API gateway

## **Getting Started**

<AccordionGroup>
  <Accordion title="Why should I use Anannas?">
    Anannas is a unified API gateway that provides access to multiple LLM providers (OpenAI, Anthropic, Google, Meta, Mistral, and more) through a single interface. Key benefits include:

    * **Unified API**: One endpoint for all models with consistent authentication and response formats
    * **Intelligent Routing**: Automatic load balancing and provider selection based on price, latency, or throughput
    * **Minimal Overhead**: Sub-20ms latency with comprehensive monitoring and analytics
    * **Cost Optimization**: Route requests to the most cost-effective providers automatically
    * **Fallback Support**: Automatic failover when preferred providers are unavailable
    * **Developer-First**: Built for developers with structured outputs, tool calling, and multimodal support
  </Accordion>

  <Accordion title="How do I get started with Anannas?">
    Getting started with Anannas is simple:

    1. **Register**: Create an account at [anannas.ai](https://anannas.ai)
    2. **Generate API Key**: Create an API key from your dashboard
    3. **Make Your First Request**: Use our OpenAI-compatible endpoints to send requests

    ```bash theme={null}
    curl -X POST "https://api.anannas.ai/v1/chat/completions" \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "openai/gpt-5-mini",
        "messages": [{"role": "user", "content": "Hello Anannas!"}]
      }'
    ```

    Check our [Quick Start Guide](/quickstart) for detailed examples.
  </Accordion>

  <Accordion title="How do I get support?">
    We provide multiple support channels:

    * **Community**: Join our [Telegram](https://t.me/+8KtRNZLaa_M4NDA1) for community support and discussions
    * **Email**: Contact us at [contact@anannas.ai](mailto:contact@anannas.ai) for technical support
    * **Documentation**: Comprehensive guides and API references in our docs
    * **Status Page**: Check system status and uptime at our status page

    For enterprise customers, we offer dedicated support with SLA guarantees.
  </Accordion>
</AccordionGroup>

## **Models and Providers**

<AccordionGroup>
  <Accordion title="What LLM models does Anannas support?">
    Anannas supports a comprehensive range of models from major AI providers:

    **Models:**

    * **OpenAI**: GPT-5, GPT-4o-mini, GPT-3.5-turbo, and more
    * **Anthropic**: Claude 4.5 Sonnet, Claude 3 Opus, Claude 3 Haiku
    * **xAI**: Grok-1, Grok-2 (via direct integration)
    * **Meta**: Llama 3.1, Llama 3.3 series (8B, 70B, 405B)
    * **Google**: Gemini Pro, Gemini Flash, PaLM models
    * **Mistral**: Mistral 7B, Mixtral 8x7B, Codestral
    * **Cohere**: Command, Command Light, Aya models
    * **Alibaba**: Qwen series (Qwen 2.5, Qwen 2.7B, 72B)
    * **DeepSeek**: DeepSeek Coder, DeepSeek Chat models
    * Many more models...

    View the complete list at our [Models page](https://anannas.ai/models).
  </Accordion>

  <Accordion title="How does provider auto-routing work?">
    Anannas includes intelligent routing that automatically selects the best provider for your requests:

    **Routing Strategies:**

    * **Price-based**: Routes to the most cost-effective provider
    * **Latency-based**: Selects providers with lowest response times
    * **Throughput-based**: Chooses providers with highest capacity
    * **Load-balanced**: Distributes requests evenly with price weighting

    **Provider Preferences:**

    ```json theme={null}
    {
      "model": "gpt-4o",
      "provider": {
        "sort": "price",
        "only": ["openai", "anthropic"],
        "ignore": ["google"],
      }
    }
    ```

    **Fallback Support:**

    * Automatic failover when primary providers are unavailable
    * Zero-downtime switching between providers
    * Real-time provider health monitoring
  </Accordion>

  <Accordion title="How frequently are new models added?">
    We continuously expand our model catalog:

    * **Weekly Updates**: New model releases are typically available within days
    * **Partnership Access**: Early access to models through provider partnerships
    * **Community Requests**: Submit model requests via our support channels
  </Accordion>
</AccordionGroup>

## **API and Integration**

<AccordionGroup>
  <Accordion title="What API formats does Anannas support?">
    Anannas provides multiple API formats for maximum compatibility:

    **Completely OpenAI-Compatible:**

    ```bash theme={null}
    POST /v1/chat/completions
    ```

    All endpoints support streaming responses and maintain consistent authentication.
  </Accordion>

  <Accordion title="Does Anannas support streaming responses?">
    Yes! Anannas supports real-time streaming for all compatible models:

    ```bash theme={null}
    curl -X POST "https://api.anannas.ai/v1/chat/completions" \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "gpt-4o",
        "messages": [{"role": "user", "content": "Tell me a story"}],
        "stream": true
      }'
    ```

    Streaming works across all providers with automatic provider selection and fallback support.
  </Accordion>

  <Accordion title="What authentication methods are supported?">
    Anannas uses API key authentication:

    **API Key Authentication:**

    ```bash theme={null}
    Authorization: Bearer YOUR_API_KEY
    ```

    **Key Management:**

    * Generate multiple API keys for different applications
    * Set usage limits and restrictions per key
    * Rotate keys without downtime
    * Monitor key usage and analytics

    **Security Features:**

    * Rate limiting per API key
    * IP allowlisting (enterprise)
    * Request signing (coming soon)
    * Audit logs for all API usage
  </Accordion>

  <Accordion title="How do I handle errors and retries?">
    Anannas provides comprehensive error handling:

    **Standard Error Format:**

    ```json theme={null}
    {
      "error": {
        "type": "invalid_request_error",
        "message": "Model not found",
        "code": "model_not_found"
      }
    }
    ```

    **Retry Logic:**

    * Automatic retries with exponential backoff
    * Provider fallback on failures
    * Circuit breaker for unhealthy providers
    * Detailed error codes for programmatic handling

    **Error Categories:**

    * `authentication_error`: Invalid API key
    * `rate_limit_error`: Rate limit exceeded
    * `insufficient_quota`: Credits exhausted
    * `model_not_found`: Unsupported model
    * `provider_error`: Provider-specific issues
  </Accordion>
</AccordionGroup>

## **Pricing and Billing**

<AccordionGroup>
  <Accordion title="How does billing work on Anannas?">
    Anannas uses a transparent credit-based billing system:

    **Credit System:**

    * Purchase credits in advance
    * Pay only for actual token usage
    * No monthly subscriptions required
    * Credits never expire

    **Pricing Structure:**

    * Model-specific pricing
    * Different rates for prompt and completion tokens
    * Special pricing for image generation and reasoning tokens
    * Volume discounts for high usage

    **Billing Features:**

    * Real-time usage tracking
    * Detailed usage analytics
    * Invoice generation
    * Multiple payment methods (Stripe)
  </Accordion>

  <Accordion title="What payment methods are accepted?">
    We accept multiple payment methods through Stripe:

    * **Credit Cards**: Visa, Mastercard, American Express
    * **Debit Cards**: All major debit card networks
    * **Digital Wallets**: Apple Pay, Google Pay (where available)
    * **Cryptocurrency**: Bitcoin, Ethereum (coming soon)
    * **UPI**: (coming soon)

    All payments are processed securely through Stripe with PCI compliance.
  </Accordion>

  <Accordion title="Are there any fees for using Anannas?">
    Our pricing is transparent with minimal fees:

    **Credit Purchase Fees:**

    * 5% processing fee on credit purchases
    * Minimum fee of \$0.80 per transaction
    * No fees for API usage beyond model costs

    **BYOK (Bring Your Own Keys):**

    * No additional fees when using your own provider API keys
    * Only pay for Anannas infrastructure and routing features
    * Full control over your provider relationships
  </Accordion>

  <Accordion title="What's the refund policy?">
    Our refund policy is developer-friendly:

    **Credit Refunds:**

    * Unused credits can be refunded within 30 days
    * Full refund for technical issues or service problems
    * Prorated refunds for subscription cancellations

    **Usage Refunds:**

    * We don't charge for any error in inference
    * or for provider-side errors
    * Dispute resolution within 48 hours

    **Enterprise Customers:**

    * Custom refund terms in enterprise agreements
    * SLA-based credit compensation
    * Dedicated support for billing issues

    Contact support for any issues.
  </Accordion>
</AccordionGroup>

## **Features and Capabilities**

<AccordionGroup>
  <Accordion title="What advanced features does Anannas support?">
    Anannas provides comprehensive AI capabilities:

    **Structured Outputs:**

    * JSON schema enforcement
    * Type-safe responses
    * Custom output formats

    **Tool Calling:**

    * Function execution through LLMs
    * Multi-turn tool conversations
    * Custom tool definitions

    **Multimodal Support:**

    * Image input processing
    * PDF document analysis
    * Video content understanding (coming soon)

    **Advanced Routing:**

    * Custom provider preferences
    * Geographic routing
    * Load balancing strategies
    * A/B testing capabilities
  </Accordion>

  <Accordion title="How does Anannas handle rate limits?">
    Our rate limiting system is designed for fairness and reliability:

    **Per-API-Key Limits:**

    * Burst allowance for short spikes
    * Automatic scaling based on usage patterns

    **Provider-Level Limits:**

    * Respects individual provider rate limits
    * Automatic load balancing across providers
    * Queue management for high-demand periods

    **Enterprise Limits:**

    * Custom rate limits based on plan
    * Priority access during high traffic
    * Dedicated capacity allocation
  </Accordion>

  <Accordion title="Does Anannas support organization accounts?">
    Yes! Organization accounts provide team management features:

    **Organization Features:**

    * Team member management
    * Role-based access control
    * Shared billing and usage tracking
    * Organization-wide API keys

    **Roles Available:**

    * **Owner**: Full access and billing control
    * **Admin**: User and billing management
    * **Member**: API access with usage monitoring
    * **Viewer**: Read-only access to analytics

    **Billing Benefits:**

    * Consolidated billing for all team usage
    * Volume discounts across team usage
    * Invoice management for accounting
    * Usage analytics per team member
  </Accordion>
</AccordionGroup>

## **Privacy and Security**

<AccordionGroup>
  <Accordion title="What data does Anannas log and store?">
    We maintain strict privacy standards:

    **Logged Data:**

    * Request metadata (timestamp, model, token counts)
    * API key usage statistics - for dashboard

    **Not Logged:**

    * Request content
    * Response content
    * Personal information

    **Data Retention:**

    * Usage logs: 90 days for analytics
    * No content storage beyond processing time
  </Accordion>

  <Accordion title="What are Zero Data Retention (ZDR) providers?">
    ZDR providers ensure no data persistence:

    **ZDR Features:**

    * No request/response logging
    * In-memory processing only
    * Automatic data deletion after processing
    * Compliance with strict data regulations

    **Use Cases:**

    * Healthcare data processing
    * Financial information handling
    * Legal document analysis
    * Sensitive corporate data

    **ZDR Providers:**

    * Select providers offer ZDR guarantees
    * Audit logs for compliance verification
  </Accordion>
</AccordionGroup>

## **Account and API Management**

<AccordionGroup>
  <Accordion title="How do I manage multiple API keys?">
    Anannas supports multiple API keys for different applications and environments:

    **Key Management Features:**

    * Create as many API keys per account
    * Set custom names for easy identification
    * Monitor usage per key in the dashboard
    * Revoke keys instantly when compromised

    **Best Practices:**

    * Use separate keys for development, staging, and production
    * Name keys descriptively (e.g., "production-web-app", "staging-mobile")
    * Rotate keys regularly for security
    * Monitor usage patterns to detect anomalies

    **Organization Keys:**

    * Organizations can create shared API keys
    * Team members can use organization keys
    * Usage is tracked per organization
    * Admins can manage all organization keys
  </Accordion>

  <Accordion title="How do I monitor my API usage and costs?">
    Comprehensive usage monitoring is available through the dashboard:

    **Real-time Analytics:**

    * Live usage dashboard with request counts
    * Token usage breakdown by model
    * Cost tracking and projections
    * Provider routing statistics

    **Usage Alerts:**

    * Low balance warnings and auto topup triggers
    * High usage notifications
    * Rate limit approaching alerts
    * Monthly usage summaries

    **Detailed Reports:**

    * Daily, weekly, monthly usage reports
    * Model performance comparisons
    * Cost optimization suggestions
    * Export data for accounting

    Access your dashboard at [anannas.ai/dashboard](https://anannas.ai/).
  </Accordion>

  <Accordion title="How do I set up team accounts and organizations?">
    Organization accounts provide team collaboration features:

    **Creating Organizations:**

    1. Go to your dashboard → Organizations
    2. Click "Create Organization"
    3. Set organization name
    4. Invite team members via email

    **Billing Benefits:**

    * Consolidated billing for all team usage
    * Volume discounts across team members
    * Shared credit pool
    * Invoice management for accounting teams

    **Invitation Process:**

    * Send invitations via email
    * Members accept invitations to join
    * Automatic role assignment
    * Usage tracking per member
  </Accordion>
</AccordionGroup>

## **Troubleshooting**

<AccordionGroup>
  <Accordion title="What should I do if my requests are failing?">
    Follow these steps to diagnose and resolve request failures:

    **Immediate Checks:**

    1. Verify API key is valid and active
    2. Check account balance and credits
    3. Confirm model name is correct
    4. Validate request format and parameters

    **Provider Issues:**

    * Check provider status in dashboard
    * Try different models or providers
    * Use provider routing to avoid failed providers
    * Monitor provider-specific error rates

    **Debugging Tools:**

    * Use request/response logging
    * Check error details in dashboard
    * Monitor provider routing decisions
    * Review usage analytics for patterns

    **Getting Help:**

    * Check our status page for outages
    * Contact support with error details
    * Join our [Discord community](https://anannas.ai/discord) for help
    * Review documentation for examples
  </Accordion>
</AccordionGroup>

<div
  style={{
marginTop: '2rem',
display: 'flex',
alignItems: 'center',
gap: '0.75rem',
fontSize: '14px',
color: '#6b7280',
}}
>
  <span>Was this page helpful?</span>

  <button
    id="feedback-yes-btn"
    onClick={() => {
  console.log('Positive feedback');
  document.getElementById('feedback-yes-btn').style.background =
    'rgba(156, 163, 175, 0.1)';
  document.getElementById('feedback-yes-btn').style.backdropFilter =
    'blur(10px)';
  document.getElementById('feedback-yes-btn').style.border =
    '1px solid rgba(156, 163, 175, 0.2)';
  document.getElementById('feedback-no-btn').style.background =
    'transparent';
  document.getElementById('feedback-no-btn').style.backdropFilter = 'none';
  document.getElementById('feedback-no-btn').style.border = 'none';
}}
    style={{
  background: 'transparent',
  border: 'none',
  cursor: 'pointer',
  padding: '6px 12px',
  fontSize: '14px',
  borderRadius: '8px',
  transition: 'all 0.2s ease',
  display: 'flex',
  alignItems: 'center',
  gap: '6px',
  color: '#6b7280',
}}
  >
    <span style={{ fontSize: '16px' }}>👍</span>
    Yes
  </button>

  <button
    id="feedback-no-btn"
    onClick={() => {
  console.log('Negative feedback');
  document.getElementById('feedback-no-btn').style.background =
    'rgba(156, 163, 175, 0.1)';
  document.getElementById('feedback-no-btn').style.backdropFilter =
    'blur(10px)';
  document.getElementById('feedback-no-btn').style.border =
    '1px solid rgba(156, 163, 175, 0.2)';
  document.getElementById('feedback-yes-btn').style.background =
    'transparent';
  document.getElementById('feedback-yes-btn').style.backdropFilter = 'none';
  document.getElementById('feedback-yes-btn').style.border = 'none';
}}
    style={{
  background: 'transparent',
  border: 'none',
  cursor: 'pointer',
  padding: '6px 12px',
  fontSize: '14px',
  borderRadius: '8px',
  transition: 'all 0.2s ease',
  display: 'flex',
  alignItems: 'center',
  gap: '6px',
  color: '#6b7280',
}}
  >
    <span style={{ fontSize: '16px' }}>👎</span>
    No
  </button>
</div>