Frequently Asked Questions¶

General Questions¶

What is the Wickson API?¶

The Wickson API is an all-in-one solution for media analysis, vector searching, and multimodal AI tooling. It processes over 50 file types, from text and images to videos and 3D models, providing developers with streamlined media analysis, vector search capabilities, and Retrieval-Augmented Generation (RAG) functionality. Built on a state-of-the-art dual-vector storage system with transparent, pay-as-you-go pricing, Wickson simplifies building next-generation AI applications.

What sets the Wickson API apart from other solutions?¶

Wickson differentiates itself through:

Unified API for all media types (no separate endpoints for different formats)
No monthly storage fees for vectors or metadata
High-performance vector search with advanced semantic capabilities
Multi-modal understanding that connects concepts across different media types
Simple, predictable pricing that scales with your usage
Developer-first approach with clear documentation and easy implementation

How can I get started with the Wickson API?¶

Getting started is straightforward:

Create an account on the Wickson API dashboard
Add funds to your account (minimum $10) to enable API operations
Retrieve your API key from the dashboard
Follow our quickstart guide to make your first API request

For a comprehensive introduction, see our Media Processing Guide.

Technical Questions¶

What file formats does Wickson support?¶

The Wickson API supports over 50 file formats across multiple categories, including:

Documents & Data - PDF, DOCX, TXT, HTML, XML, JSON, CSV/TSV, and more

Images - JPEG, PNG, GIF, BMP, TIFF, WebP, SVG, HEIC/HEIF, and more

Video & Audio - MP4, MOV, AVI, WebM, MPEG, MP3, WAV, and more

3D Models - STL, OBJ, GLB, 3MF, PLY, DAE, and limited FBX support

Office & Productivity - XLSX, PPTX, ODP, ODS, and more

For detailed information, see our Media Types Guide.

What are the file size limits?¶

Current file size limits are:

Documents: 20MB / 128K tokens
Images: 20MB
Video: 100MB / 10 minutes
Audio: 50MB / 20 minutes
3D Models: 50MB

These limits may be adjusted during the beta phase as we optimize our processing pipelines.

The Wickson API's search capabilities go beyond simple keyword matching by understanding the actual meaning of content across different media types. Our advanced semantic search can find conceptually similar content even when the exact terms aren't present.

For example, searching for "happy moments" could return:

Images of smiling people
Videos with celebrations
Documents describing positive experiences
Audio files with joyful sounds

This is possible because all media is converted to vector embeddings that capture semantic meaning, allowing the system to find relationships based on concepts rather than just keywords. For implementation details, see our Search Guide.

What's the difference between basic and advanced search?¶

The Wickson API offers two search approaches:

Basic Search (FREE)

Optimized for speed and direct relevance
Searches for immediate semantic matches
Best for straightforward queries with clear intent
Returns results ordered by similarity score

Advanced Search ($0.01 + $0.01 per depth level)

Uses our R3F search technology
Explores contextual connections between content
Discovers cross-modal relationships and deeper associations
Best for research, exploration, and discovering unexpected connections
Returns results with explanation of relationships discovered

See our Search Guide for more details.

How do I implement RAG (Retrieval-Augmented Generation) with Wickson?¶

The Wickson API provides robust tooling for RAG implementations through:

Media Processing: Upload and process your content library to create vector embeddings
Vector Search: Use semantic search to retrieve relevant content based on queries
Analysis Endpoints: Extract insights from media without permanent storage

For effective RAG implementations, we recommend:

Organizing related content in collections
Using batch processing for efficient content ingestion
Implementing advanced search for comprehensive retrieval

How do collections work?¶

Collections are logical groupings that organize your media items. They help structure your content library and optimize search performance. Key features include:

Media items can belong to one collection at a time
Collections are created automatically when you specify a new collection_id
You can move items between collections at any time
Collections can be used to target specific content in search operations
There's no limit to the number of collections you can create

The default collection is created automatically and holds all media not explicitly assigned elsewhere.

What happens to my original files after processing?¶

The Wickson API is designed to extract meaning, not store files. When you upload content:

Your file is temporarily stored during the processing stage
We extract vector embeddings and rich metadata (summaries, entities, topics, etc.)
The original file is securely deleted from our systems
Only the derived data (vectors and metadata) is retained and stored in the vector database for searching

This approach eliminates redundant storage costs, keeps your data architecture clean, and enhances security and data ownership - You maintain control of your original files in your preferred storage system.

How can I integrate Wickson with my existing storage solution?¶

The Wickson API is designed to help you build next-generation applications as well as complement your existing infrastructure. By pairing Wickson with a traditional file storage service (like AWS S3, Google Drive, or internal systems or storage), you can easily create solutions that not only search and retrieve results, but recall or interact with the original files you have stored as well.

Common integration patterns include:

Process-then-Search: Upload files to Wickson for processing, search by meaning, then retrieve matching originals from your storage using the file identifiers
Webhook Processing: Configure your storage system to trigger Wickson processing when new files are added
Batch Synchronization: Periodically scan your storage for new content and process it in batches

If Wickson doesn't store my original files, how do I retrieve the content I find in search results?¶

When you perform searches, the Wickson API returns:

Unique identifiers for each matching item
Rich metadata about the content (summary, entities, topics, etc.)
Content extracts relevant to your query
Vector embeddings (when requested)

To access the full original file, you'll need to maintain a mapping between Wickson's item IDs and your files in your storage system. This is typically done by:

Storing the Wickson media ID with your file metadata
Using consistent file naming/paths that can be reconstructed
Creating a lookup table in your application database

This separation of concerns keeps your architecture clean and gives you complete control over your original content.

Can I use Wickson if I don't want to maintain my own file storage?¶

Absolutely! While Wickson doesn't store your original files, there are several approaches you can take:

Content extracts may be sufficient: For many use cases, the detailed summaries, entity information, and content extracts provided by Wickson contain all the information you need
Cloud storage integration: Use a cloud storage solution like AWS S3, Google Cloud Storage, or Azure Blob Storage alongside Wickson
Hybrid approach: Store only your most important files externally, while using Wickson's content extracts for the rest

What information can I expect in search results if the original files aren't stored?¶

While original files aren't stored, Wickson's search results are incredibly rich:

Comprehensive summaries capturing the essence of the content
Key entities like people, organizations, locations, and concepts
Topic classification and thematic analysis
Sentiment and emotional analysis where applicable
Content extracts showing the most relevant portions
Contextual relationships between different pieces of content (in advanced search)

For most use cases, this derived information provides all the insights needed without requiring access to the original file.

Pricing and Billing¶

What is the pricing model for the Wickson API?¶

The Wickson API uses transparent, pay-as-you-go pricing with no monthly storage fees:

Operation	Cost	Notes
Data Storage	FREE	All vector embeddings and metadata
Media Processing	$0.03 + $0.01 one-time I/O charge	Per file (any media type)
Media Analysis	$0.03	Per query
LLM Query	$0.01	Per query plus context length charges
Basic Search	FREE	Per basic hybrid search
Similarity Search	$0.01	Per similarity search
Advanced Search	$0.01 + $0.01 per depth	Per advanced hybrid search plus depth charge

You only pay for the operations you perform, with no recurring charges for stored content.

Are there any monthly or minimum fees?¶

No. The Wickson API has:

No monthly subscription fees
No minimum usage requirements
No ongoing storage charges
No hidden costs

You only pay for the specific operations you perform, making costs predictable and aligned with your actual usage.

How do I monitor my usage and costs?¶

You can track your usage and costs through:

Dashboard: Log in to app.wickson.ai to view detailed usage statistics and transaction history
API Endpoint: Use the /v1/usage endpoint to programmatically retrieve usage data
Response Headers: Each API response includes your current balance in the X-Account-Balance header

Security¶

How is my data secured with the Wickson API?¶

Security is a top priority for the Wickson API:

Data Encryption: All data is encrypted both in transit and at rest
API Key Security: API keys use strong encryption and can be regenerated at any time
Isolated Processing: Your content is processed in isolated environments
No Data Retention: Original files are deleted after processing is complete
Your Control: You can delete your vectors and metadata at any time

Does Wickson use my data to train AI models?¶

No. We do not use your data to train our models or for any purpose other than providing the service you've requested. Your data remains your property at all times. For more about this please visit

We encourage you to check out our privacy policy and terms of service for more information

Can I delete my data?¶

Yes. You can delete individual media items, entire collections, or all of your data at any time through:

API Endpoints: Use our deletion endpoints to programmatically remove content
Dashboard: Get statistics and information about uploaded content through the web interface, or request account and data deletion on the accounts page

Deletion is permanent and removes all associated vectors and metadata from our systems.

Support and Resources¶

Where can I find more documentation?¶

Our comprehensive documentation is available at:

API Reference: Detailed endpoint specifications
Guides: Best practices and implementation strategies
Getting Started: Tutorials for new users
Media Types: Information about supported formats

How can I get help with implementation?¶

If you need assistance implementing the Wickson API:

Support System: Log in to your dashboard at app.wickson.ai and use the support tab
Documentation: Explore our detailed guides and reference materials
Email: For non-technical inquiries, contact sales@firespawnstudios.net

See our Contact Support page for more details.

Is there a community forum or discussion group?¶

We're in the process of establishing community resources. For now, the best way to connect is through our support channels. Stay tuned for announcements about community forums and discussion groups!

Additional technical details¶

For LLM Query operations that leverage Gemini models, context length charges apply based on token usage. These rates generally follow Google's Gemini Flash pricing structure. For current rates and detailed information, see Google's Gemini API Pricing. Firespawn Studios and the Wickson API is not affiliated with, endorsed by, or sponsored by Google. Google and Gemini are trademarks of Google LLC.