Trieve website screenshot

Trieve

Trieve (Devflow, Inc.) is an open-source, all-in-one search, recommendations, RAG, and analytics platform delivered as a REST API. The backend is written in Rust (Actix-web) and exposes endpoints for managing organizations, datasets, chunks, chunk groups, files, search, topics / messages (LLM chat), web crawls, events, analytics, and experiments. Trieve Cloud is hosted at api.trieve.ai; the same server can be self-hosted from the devflowinc/trieve repository. Official SDKs are published for TypeScript and Python.

16 APIs 0 Features
SearchRAGVector SearchHybrid SearchRecommendationsAnalyticsOpen Source

APIs

Trieve REST API

Production REST API for Trieve Cloud. Organizes resources around organizations, datasets, chunks, chunk groups, files, search, topics, messages, crawls, analytics, events, exper...

Trieve Chunk API

Manage chunks - the individual searchable units of content stored in a dataset - including create, update, delete, get, and bulk operations.

Trieve Chunk Group API

Group chunks into bookmark-style folders for organization, recommendations, and group-scoped search within a dataset.

Trieve Dataset API

Create and configure datasets that hold chunks, chunk groups, and search / RAG configuration for a workload.

Trieve Search API

Run vector, full-text, and hybrid search across a dataset, with filters, boosts, re-ranking, and highlights.

Trieve File API

Upload files (up to 1 GB), extract text, and chunk them into a dataset. Asynchronous - completion is signaled via the Events API.

Trieve Topic API

Topics persist conversational state for generative-AI chat sessions backed by a Trieve dataset.

Trieve Message API

Messages are turns within a topic; the Message API drives RAG completions, streaming responses, and citation-aware chat against a dataset.

Trieve Crawl API

Create and manage web crawls that ingest pages into a dataset for search and RAG.

Trieve Analytics API

Query search and RAG analytics - top queries, no-result queries, latency, click-through, and experiment outcomes.

Trieve Events API

Server-sent notifications about asynchronous work such as file processing, crawl completion, and chunk ingestion.

Trieve Organization API

Manage organizations, roles, and members.

Trieve Auth API

Registration, login, and session handling for Trieve users.

Trieve TypeScript SDK

Official TypeScript / JavaScript client library for the Trieve REST API.

Trieve Python SDK

Official Python client library (trieve-py-client) for the Trieve REST API.

Trieve Open Source Server

Self-hostable Trieve server (Rust / Actix-web), dashboard, search and chat UIs, ingestion / file / delete workers, batch-ETL utilities, pdf2md converter, and Helm charts for Kub...

Collections

Trieve REST API

OPEN

Pricing Plans

Trieve Plans Pricing

1 plans

PLANS

Rate Limits

Trieve Rate Limits

2 limits

RATE LIMITS

FinOps

Trieve Finops

FINOPS

Resources

🔗
DomainSecurity
DomainSecurity
🔑
Authentication
Authentication
🔗
Website
Website
🔗
Documentation
Documentation
👥
GitHub
GitHub
💰
Pricing
Pricing
📰
Blog
Blog
🔗
LlmsText
LlmsText

Sources

Raw ↑ 
opencollection: 1.0.0
info:
  name: Trieve REST API
  version: '1'
request:
  auth:
    type: apikey
    key: Authorization
    value: '{{Authorization}}'
    placement: header
items:
- info:
    name: Chunk
    type: folder
  items:
  - info:
      name: Create or upsert chunk(s)
      type: http
    http:
      method: POST
      url: https://api.trieve.ai/api/chunk
      headers:
      - name: TR-Dataset
        value: ''
    docs: 'Creates new chunks. A chunk with the same tracking_id as an existing chunk

      will cause the request to fail. Bulk uploads are capped at 120 chunks

      per request.

      '
- info:
    name: Dataset
    type: folder
  items:
  - info:
      name: Create a dataset
      type: http
    http:
      method: POST
      url: https://api.trieve.ai/api/dataset
    docs: 'Create a new dataset within an organization. Caller must be an

      organization owner.

      '
- info:
    name: File
    type: folder
  items:
  - info:
      name: Upload a file
      type: http
    http:
      method: POST
      url: https://api.trieve.ai/api/file
      headers:
      - name: TR-Dataset
        value: ''
    docs: 'Upload a file to the S3 bucket attached to a dataset. Supports naive

      chunking via Apache Tika or vision-LLM markdown conversion.

      '
- info:
    name: Topic
    type: folder
  items:
  - info:
      name: Create a chat topic
      type: http
    http:
      method: POST
      url: https://api.trieve.ai/api/topic
      headers:
      - name: TR-Dataset
        value: ''
    docs: 'Create a new chat topic. Topics coordinate conversation message history

      of gen-AI chat sessions and are attached to an owner_id.

      '
- info:
    name: Message
    type: folder
  items:
  - info:
      name: Create a message
      type: http
    http:
      method: POST
      url: https://api.trieve.ai/api/message
      headers:
      - name: TR-Dataset
        value: ''
    docs: 'Create a message attached to a topic to coordinate memory of gen-AI

      chat sessions.

      '
bundled: true