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

# FAQ

> Frequently asked questions about FLTR

# Frequently Asked Questions

Common questions and answers about using FLTR.

## General

<AccordionGroup>
  <Accordion title="What is FLTR?">
    FLTR is a Context as a Service platform that makes your documents AI-ready. It provides semantic search over your content using hybrid vector + keyword search, accessible via REST API or Model Context Protocol (MCP).
  </Accordion>

  <Accordion title="Who is FLTR for?">
    FLTR serves three main audiences:

    * **AI Developers** - Building RAG applications, knowledge bases, and AI agents
    * **No-Code Builders** - Integrating with Zapier, Make, n8n
    * **Enterprises** - Team collaboration with advanced security
  </Accordion>

  <Accordion title="How does FLTR differ from other search solutions?">
    FLTR combines:

    * **Hybrid search** - Vector similarity + keyword matching
    * **MCP-native** - Built for Claude Desktop, VS Code, Cursor
    * **Simple API** - RESTful with comprehensive docs
    * **Multimodal** - PDFs, images, text, code
  </Accordion>

  <Accordion title="Is FLTR open source?">
    The FLTR API is proprietary, but we provide:

    * Open API specification (OpenAPI 3.1)
    * Public documentation
    * Example integrations on GitHub
    * MCP server implementations
  </Accordion>
</AccordionGroup>

## Pricing & Billing

<AccordionGroup>
  <Accordion title="Is there a free tier?">
    Yes! Free tier includes:

    * 50 requests/hour (anonymous)
    * 2 datasets
    * 100 documents
    * 1GB storage

    Paid plans start at \$29/month.
  </Accordion>

  <Accordion title="How does billing work?">
    Billing is credit-based:

    * **Queries**: 1 credit per query
    * **Uploads**: 10 credits per document
    * **Storage**: 1 credit per GB per month
    * **Reranking**: +2 credits per query

    Plans include monthly credit allocations.
  </Accordion>

  <Accordion title="Can I upgrade or downgrade my plan?">
    Yes, anytime. Changes take effect immediately with prorated billing.
  </Accordion>

  <Accordion title="What payment methods do you accept?">
    We accept:

    * Credit/debit cards (Stripe)
    * ACH (US only)
    * Wire transfer (Enterprise)
    * Annual prepayment (10% discount)
  </Accordion>
</AccordionGroup>

## Technical

<AccordionGroup>
  <Accordion title="What file types are supported?">
    We support:

    * **Documents**: PDF, DOCX, PPTX, TXT, MD
    * **Images**: JPG, PNG (with OCR)
    * **Code**: PY, JS, JAVA, etc.
    * **Data**: JSON, XML, CSV, YAML

    Max file size: 10MB
  </Accordion>

  <Accordion title="How long does document processing take?">
    Processing time depends on file size:

    * **Text (\< 1MB)**: 1-5 seconds
    * **PDFs (1-5MB)**: 5-15 seconds
    * **Large PDFs (5-10MB)**: 15-30 seconds
    * **Images**: +5-10 seconds for OCR
  </Accordion>

  <Accordion title="What embedding model do you use?">
    Default: `text-embedding-3-small` (OpenAI)

    Coming soon:

    * `text-embedding-3-large`
    * Custom models (Enterprise)
    * Multilingual models
  </Accordion>

  <Accordion title="Can I use my own embedding model?">
    Not yet. Custom embedding models are planned for Enterprise plans. [Contact sales](mailto:sales@fltr.com) to discuss.
  </Accordion>

  <Accordion title="How accurate is the search?">
    Hybrid search provides:

    * **Precision**: 85-95% for exact matches
    * **Recall**: 80-90% for semantic queries
    * **With reranking**: +5-10% improvement

    Accuracy depends on document quality and metadata.
  </Accordion>

  <Accordion title="Can I search across multiple datasets?">
    Not directly. You need to query each dataset separately. Use batch queries to search multiple datasets efficiently.

    Multi-dataset search is on our roadmap for future release.
  </Accordion>
</AccordionGroup>

## Security & Privacy

<AccordionGroup>
  <Accordion title="Is my data secure?">
    Yes. FLTR implements:

    * **Encryption**: TLS 1.3 in transit, AES-256 at rest
    * **Isolation**: Data separated by account
    * **Access control**: Role-based permissions
    * **Compliance**: SOC 2 Type II (in progress)
  </Accordion>

  <Accordion title="Who can access my documents?">
    Only you and users you grant access to. FLTR staff cannot access your documents unless you explicitly grant support access for troubleshooting.
  </Accordion>

  <Accordion title="Where is data stored?">
    Data is stored in:

    * **US East (primary)**: us-east-1 (AWS)
    * **EU (optional)**: eu-west-1 (AWS)

    Enterprise plans can choose region.
  </Accordion>

  <Accordion title="Do you train AI models on my data?">
    **No.** Your data is never used for:

    * Model training
    * Research
    * Analytics (except aggregated usage stats)
    * Any purpose outside your use case
  </Accordion>

  <Accordion title="Can I delete my data?">
    Yes. You can:

    * Delete documents anytime
    * Delete datasets (removes all documents)
    * Delete account (removes all data within 30 days)

    Backups are retained for 30 days, then permanently deleted.
  </Accordion>

  <Accordion title="Are you GDPR compliant?">
    Yes. FLTR is GDPR compliant with:

    * Data processing agreements
    * Right to deletion
    * Data portability
    * Privacy by design
  </Accordion>
</AccordionGroup>

## Integration

<AccordionGroup>
  <Accordion title="How do I integrate FLTR with my app?">
    Three ways:

    1. **REST API** - For any language
    2. **MCP** - For Claude Desktop, VS Code, Cursor
    3. **No-code** - Zapier, Make, n8n

    See [Getting Started](/quickstart/getting-started).
  </Accordion>

  <Accordion title="Is there a Python SDK?">
    Not yet. We're working on official SDKs for:

    * Python
    * JavaScript/TypeScript
    * Go

    For now, use the REST API directly.
  </Accordion>

  <Accordion title="Can I use FLTR with Claude Desktop?">
    Yes! FLTR has native MCP support for Claude Desktop.

    See [OAuth Setup](/authentication/oauth) for configuration.
  </Accordion>

  <Accordion title="Does FLTR work with LangChain?">
    Yes, use FLTR as a retriever in LangChain:

    ```python theme={null}
    from langchain.retrievers import FLTRRetriever

    retriever = FLTRRetriever(
        api_key="your_key",
        dataset_id="ds_abc123"
    )
    ```

    See our [LangChain integration guide](/integrations/langchain).
  </Accordion>

  <Accordion title="Can I self-host FLTR?">
    No. FLTR is a managed service only.

    Enterprise plans include:

    * Dedicated infrastructure
    * Custom domains
    * VPC peering
  </Accordion>
</AccordionGroup>

## Support

<AccordionGroup>
  <Accordion title="What support do you offer?">
    Support varies by plan:

    * **Free**: Email support (48h response)
    * **Pro**: Email + chat (24h response)
    * **Enterprise**: Priority support + Slack (4h response)
  </Accordion>

  <Accordion title="Do you have a status page?">
    Coming soon. We're setting up a public status page for real-time service health monitoring.

    It will include:

    * Email notifications
    * SMS alerts
    * Slack integration
  </Accordion>

  <Accordion title="How do I report a bug?">
    Report bugs via:

    * Email: [support@fltr.com](mailto:support@fltr.com)

    Include:

    * Steps to reproduce
    * Expected vs actual behavior
    * Request/response examples
  </Accordion>

  <Accordion title="Can I request features?">
    Yes! We love feedback. Request features via:

    * Email: [support@fltr.com](mailto:support@fltr.com)

    We prioritize based on user demand.
  </Accordion>
</AccordionGroup>

## Still Have Questions?

<CardGroup cols={2}>
  <Card title="Email Support" icon="envelope" href="mailto:support@fltr.com">
    [support@fltr.com](mailto:support@fltr.com)
  </Card>

  <Card title="Documentation" icon="book" href="/introduction">
    Browse complete docs
  </Card>

  <Card title="Contact Sales" icon="calendar" href="mailto:sales@fltr.com">
    Schedule a demo
  </Card>

  <Card title="Troubleshooting" icon="wrench" href="/support/troubleshooting">
    Common issues and fixes
  </Card>
</CardGroup>