| What it does | Open-Source Hybrid Search System
| Personal News Reader with Intelligence Training
|
|---|
| Stage | Not given
| Established
|
|---|
| Founded | Not given
| 2009
|
|---|
| Based in | Not given
| US
|
|---|
| Pricing model | Not given
| freemium
|
|---|
| Pricing | Not given
| Free plan with up to 64 sites. Premium $36/year for 1,024 sites, search, and river of news. Premium Archive $99/year for full archive, Ask AI, and 4,096 sites. Premium Pro $29/month for 10,000 sites and 5-minute feed fetching. All new accounts include a 30-day free premium trial with no credit card
|
|---|
| Key features | - Zero glue code with internal handling of queueing, deduplication, distributed locking, retries, and embeddings
- Server-side fusion for dense vectors, sparse models, and keyword tokens
- Built-in dashboard and real-time metrics endpoints for monitoring
- Scales from single container to distributed architecture
- Autonomous MLOps with self-orchestrating encoder workers
- Works with PostgreSQL, MariaDB, or Qdrant for storage
- WMTR (Weighted Multilevel Token Representation) for handling messy enterprise text and identifiers
- Typeahead-level latency on large corpora
| - Intelligence Training - Train stories by author, tag, title, and keyword to highlight or hide matching stories
- Full-Text Training - Train on the actual content of stories to filter based on what's being said
- Ask AI - Ask questions about stories using Claude, GPT, Gemini, or Grok
- Daily Briefing - AI-generated summary of top stories delivered daily, customizable by interests
- Web Feeds - Follow any website even without RSS by generating custom feeds
- Full-Text Search - Elasticsearch-powered search across all subscriptions and saved stories
- Story Archive - Every story archived and searchable forever with complete feed backfill history
- Native iOS and Android Apps - First-class native apps built specifically for each platform
- Real-Time Updates - Stories pushed as they publish with real-time notifications
- Story Clustering - Group duplicate stories across feeds using semantic duplicate detection
|
|---|
| What makes it different | - Unified REST API eliminates need to integrate separate embedding, queue, vector database, and fusion components
- Combines keyword precision and semantic relevance with WMTR for superior handling of enterprise identifiers and part numbers
- Flexible storage backend supporting PostgreSQL, MariaDB, or Qdrant with consistent API
- Self-orchestrating encoder workers eliminate manual model pinning to machines
| - Intelligence Training lets you control what you see by training on authors, tags, titles, and full text
- AI-powered features including Ask AI and Daily Briefing with multiple model options
- Web Feeds capability to follow any website without requiring RSS feeds
- Independent, sustainable business model charging for premium accounts since 2010
|
|---|
| Free plan | Not given
| Not given
|
|---|
| Open source | Not given
| Not given
|
|---|
| Platforms | Not given
| Not given
|
|---|
| Public API | Not given
| Not given
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 3 ● no change since last check | 75 ● no change since last check |
|---|
| Trust Flow (Majestic) | 22 | 59 |
|---|
| Citation Flow (Majestic) | 27 | 53 |
|---|
| Referring domains (Majestic) | 70 | 5417 |
|---|