Replace your dedicated engineering team with a reliable, scalable solution.
Instead of dedicating engineering resources to build and maintain your own infrastructure, use Mixpeek's battle-tested solution.
Everything you need in two API calls, instead of months of development.
Focus your team on building features, not maintaining infrastructure.
Fixed monthly pricing instead of variable infrastructure costs.
Compare the total cost of ownership between building in-house and using Mixpeek.
If you don't see an answer to your question, you can send us an email from our contact form.
Namespaces are isolated environments for your content. Each namespace can have its own collections, feature extractors, and search configurations. They're useful for separating different applications or use cases within your account. For example, you might have one namespace for product documentation and another for customer support content.
Features are extracted data points and characteristics from your content that enable advanced search and analysis capabilities. Our AI models automatically extract different types of features depending on your content:
Feature extractors are AI models that analyze your content to understand its meaning and context. They create vector embeddings (numerical representations) of your content that enable semantic search. Vector indexes organize these embeddings for efficient similarity search. Mixpeek automatically handles model selection and index optimization based on your content type.
Custom feeds are automated data pipelines that can ingest content from your specific data sources. While standard feeds support common sources like S3 and web URLs, custom feeds (available in Enterprise) can integrate with proprietary systems, custom databases, or specialized content management systems. This includes real-time syncing and custom preprocessing steps.
Queries can be text, images, or mixed media. Our system combines vector similarity search with traditional filters and ranking methods. Query limits are measured per minute and apply to API calls for searching. Advanced features like reranking and personalization are available to optimize results. The system automatically handles cross-modal search (e.g., finding images with text queries).
SLAs cover system availability, query response times, and processing throughput. Standard tier (99.9%) ensures monthly uptime and basic performance. Enhanced SLA (99.95%) adds guaranteed processing times and faster support response. Enterprise SLAs (99.99%) can include custom performance metrics, dedicated infrastructure, and guaranteed resource allocation for your specific needs.