Trieve

AI search infrastructure combining language models

  • Revolutionizes application search with AI and human fine-tuning

  • Enhances search accuracy with SPLADE and semantic vector search

  • Supports comprehensive search strategies with hybrid search and relevance tuning

Pricing:

💲 Freemium

Categories:

#Data & Research

What is Trieve

Trieve is an AI search infrastructure designed to integrate advanced search capabilities into applications. It combines powerful language models with human fine-tuning tools, offering features like semantic vector search, SPLADE full-text neural search, and hybrid search models. Aimed at developers and businesses looking to enhance their search functionality, Trieve provides a flexible, production-ready solution that can adapt to various data types and search requirements.

Key Features

  • Private Managed Embedding Models: Users can bring their own embedding model or choose from Trieve’s open-source hosted options.
  • SPLADE Full-Text Neural Search: A state-of-the-art retrieval model that surpasses traditional BM25 for full-text search.
  • Semantic Vector Search: Supports out-of-the-box semantic vector search for more nuanced query responses.
  • Hybrid Search: Combines full-text and semantic vector search with cross-encoder re-ranker models for comprehensive search results.
  • Merchandising Relevance Tuning: Allows boosting of search results based on sales or popularity.
  • Date Recency Biasing: Ensures the most recent results are shown first by supporting date recency biasing.
  • Duplicate Content Detection and Merging: Novel approach to identifying and merging duplicate content within knowledge management.
  • Sub-Sentence Highlighting: Highlights specific parts of search results to directly address user queries.
  • Open Source with Business Source License: While not free software, Trieve is open source under certain restrictions, encouraging community contributions.

Pricing

DEV Plan

  • Cost: $25 per month
  • Features:
    • 100,000 vectorized and stored chunks
    • Unlimited search queries at 10/s rate limit
    • 2 datasets
    • 5 users
    • 10,000 gen AI messages
    • 1GB S3 file storage

PRO Plan

  • Cost: $500 per month (starting)
  • Features:
    • SLO add-on available (300ms 99th percentile and 75ms 50th percentile latency)
    • SLA add-on available (99.9% uptime)
    • SOC 2, ISO 27001, and HIPAA compliance
    • 1,000,000 vectorized and stored chunks
    • Unlimited search queries at 25/s rate limit
    • 10 datasets
    • 20 users
    • 10,000 gen AI messages
    • 50GB S3 file storage

SOVEREIGN Plan

  • Cost: $1000 per month (starting)
  • Features:
    • Terraform templates
    • Helm charts
    • Run securely on your own servers or cloud
    • Multi-tenant capability for use by multiple teams within an organization
    • Unlimited usage (except for reselling)
    • Dedicated support contracts available

Please check here for more details about Trieve’s pricing.

Trieve

AI search infrastructure combining language models

Key Features

💲 Freemium

Product Embed

Subscribe to our Newsletter

Get the latest updates directly to your inbox.

Share This Tool

Related Tools

Avanty
💰 Paid
🧩 Browser Extension
⏳ Free Trial

AI-powered Chrome extension for Metabase SQL query efficiency