Search

More Filters

Show More
Add Listing
Go Back

perplexideez

0
New
Web & Internet Agents
Report

Report Abuse

Basic Information

Perplexideez is a self-hosted, AI-powered web search application intended to replicate and extend Perplexity-style search while integrating with other self-hosted services and supporting multiple users. The app uses a Postgres database for storage and can connect to either Ollama or OpenAI-compatible LLM endpoints. It performs web searches via a SearXNG instance and surfaces LLM-generated summaries with explicit source annotations so users can inspect where conclusions came from. The project supplies container images for the app and a migration utility, example Docker Compose files, Kubernetes manifest guidance, and a .env-driven configuration for SSO, AI provider selection, model mapping, and deployment settings. The repo includes local development steps, database migration commands, and a built stack using SvelteKit, Tailwind, tRPC, Prisma, AuthJS, TanStack Query, and Langchain.

Links

Stars
458
App URL
https://github.com/brunostjohn/perplexideez
Github Repository
https://github.com/brunostjohn/perplexideez/blob/main/README.md

App Details

Features
Perplexideez provides web search with LLM-generated answers plus hoverable source annotations that link back to original results. It can generate follow-up questions automatically, lets users save favourite searches, and exposes a model picker to assign different LLMs for Speed, Balanced, Quality, embeddings, titles, emoji, and media search tasks. Multi-user support and OIDC SSO are built in, with options to disable sign-up or password login. Search results and sessions can be shared via configurable links with access control and public embeds. Deployment-focused features include non-root container images, a migrate image for DB setup, Docker Compose examples, Kubernetes manifest examples, and many environment variables for AI provider, SearXNG URL, database, logging and security.
Use Cases
Perplexideez helps teams and self-hosting users run an LLM-backed search experience that integrates with existing infrastructure and enforces per-instance configuration. Multi-user accounts with separated data and OIDC SSO support make it suitable for organizations that need access control. Source annotations reduce the risk of accepting hallucinated answers by linking generated conclusions to original documents. Model selection and environment-driven configuration allow operators to balance cost and quality by choosing different models for speed, quality or embeddings. Container images and example Compose manifests simplify deployment, and the migrate image handles database migrations. The UI supports favourites, sharing with access controls, and attractive embeds to make discovered information easier to curate and distribute.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

fuji web

0 (0 review)
listing image

AutoScraper

0 (0 review)
listing image

gpt assistant

0 (0 review)
listing image

jobber

0 (0 review)
listing image

socks5 for serv00

0 (0 review)

Featured Listings

Tags

Please fill the required fields*