Search documentation

Search query

Esc

Type to search this documentation.

Home
Guides
Reference
Examples
Integrations
Changelog
Troubleshooting
API Reference
Release Notes

- Get started
- The knowledge engine
  - How Nexus works
  - Context design
    Overview
    Design your own manifest
    Configure artifact formats
  - Curation
  - Nexus model guidance
  - Queries and sessions
    Queries
    Query tracing
- Tools
  - MCP server
- Bring Your Own Cloud
- Build an assistant
  - Create an assistant
  - Manage assistants
- Upload your data
- Chat with an assistant
  - Standard interface
  - OpenAI-compatible interface
- Evaluate answers
  - Overview
  - Evaluate answers
- Retrieve context snippets
  - Overview
  - Retrieve context snippets
- Integrate with AI agents
  - Use an MCP server
- Core concepts
  - Key terms
  - Architecture
- Index data
  - Overview
  - Create and configure
    Create an index
    Configure metadata indexing
  - Data modeling
  - Data ingestion
    Overview
    Upsert records
    Import records
    Migrate from pgvector
    Check data freshness
  - Multitenancy
    Design
    Implement
  - Dedicated read nodes
    Overview
    Concepts
    Create
    Migrate
    Size and test
    Tune queries
    Scale
    Manage
  - Adopt the Documents API
  - Overview
    Overview
    Query syntax
    Text processing
  - Semantic search
    Overview
    Use a single index
    Use separate indexes
    Reciprocal rank fusion
  - Sparse-vector search
  - Filter by metadata
  - Rerank results
- Optimize
- Manage data
  - Target an index
  - Manage indexes
  - Manage namespaces
  - Manage backups
    Backups overview
    Back up an index
    Restore an index
  - Update data
  - Delete data
  - Fetch data
  - List IDs
- Manage cost
- Move to production
- Admin
- Operations
  - Integrate with cloud storage
    Amazon S3
    Google Cloud Storage (GCS)
    Azure Blob Storage
    Manage storage integrations
  - Integrate with AI agents
    MCP server
    Build a knowledge retrieval agent
  - Local development
- Using pods

Powered by Doctorine

Queries and sessions

View as Markdown Open in ChatGPT Open in ClaudeConnect to Cursor Connect to VS Code

Nexus queries

Query tracing

Was this helpful?

Suggest an edit

Propose a replacement for this page. The site team reviews it before applying any changes.

Replacement page (Markdown)Reason (optional)

Export

Documentation menu

- Get started
- The knowledge engine
  - How Nexus works
  - Context design
    Overview
    Design your own manifest
    Configure artifact formats
  - Curation
  - Nexus model guidance
  - Queries and sessions
    Queries
    Query tracing
- Tools
  - MCP server
- Bring Your Own Cloud
- Build an assistant
  - Create an assistant
  - Manage assistants
- Upload your data
- Chat with an assistant
  - Standard interface
  - OpenAI-compatible interface
- Evaluate answers
  - Overview
  - Evaluate answers
- Retrieve context snippets
  - Overview
  - Retrieve context snippets
- Integrate with AI agents
  - Use an MCP server
- Core concepts
  - Key terms
  - Architecture
- Index data
  - Overview
  - Create and configure
    Create an index
    Configure metadata indexing
  - Data modeling
  - Data ingestion
    Overview
    Upsert records
    Import records
    Migrate from pgvector
    Check data freshness
  - Multitenancy
    Design
    Implement
  - Dedicated read nodes
    Overview
    Concepts
    Create
    Migrate
    Size and test
    Tune queries
    Scale
    Manage
  - Adopt the Documents API
  - Overview
    Overview
    Query syntax
    Text processing
  - Semantic search
    Overview
    Use a single index
    Use separate indexes
    Reciprocal rank fusion
  - Sparse-vector search
  - Filter by metadata
  - Rerank results
- Optimize
- Manage data
  - Target an index
  - Manage indexes
  - Manage namespaces
  - Manage backups
    Backups overview
    Back up an index
    Restore an index
  - Update data
  - Delete data
  - Fetch data
  - List IDs
- Manage cost
- Move to production
- Admin
- Operations
  - Integrate with cloud storage
    Amazon S3
    Google Cloud Storage (GCS)
    Azure Blob Storage
    Manage storage integrations
  - Integrate with AI agents
    MCP server
    Build a knowledge retrieval agent
  - Local development
- Using pods

Powered by Doctorine