Khoj
A self-hostable personal AI and knowledge assistant with document search, agents, automations, and research workflows.
By Khoj AI · reviewed 2026-08-15
What Khoj does
Khoj combines private knowledge retrieval, local or hosted models, custom agents, scheduled automations, and access through several user interfaces. It fits people building a persistent personal assistant rather than only a document chatbot.
Pricing model
Free open source self-hosting; paid cloud available
Position in the stack
Where Khoj fits
Khoj belongs primarily in the knowledge rag layer. Khoj combines private knowledge retrieval, local or hosted models, custom agents, scheduled automations, and access through several user interfaces. It fits people building a persistent personal assistant rather than only a document chatbot. It should be evaluated as one part of a complete stack, because model files, inference providers, storage, identity, and external integrations remain separate operational choices.
Typical deployment flow
A practical sequence for evaluating Khoj before making it part of a permanent stack.
- STEP 1
Confirm the deployment boundary
Choose among desktop-local, self-hosted, managed-cloud based on users, data sensitivity, network access, and who will operate updates.
- STEP 2
Validate the complete stack
Check Desktop application, Python package, Docker, Managed cloud, connected providers, supported formats, and exact hardware or accelerator compatibility before rollout.
- STEP 3
Run a representative workflow
Test a private personal knowledge assistant, record versions and settings, then review security, backups, observability, and failure recovery.
Best for
- A private personal knowledge assistant
- Scheduled research and automations
- Combining local documents with web research
Not the right layer for
- High-throughput model serving
- A minimal inference-only deployment
Capabilities
Capabilities refer to the tool's application layer. Hardware and model support can still depend on a connected inference engine.
Deployment and compatibility
- Deployment
- Desktop / localSelf-hostedManaged cloud
- Platforms
- macOS · Windows · Linux · Docker · Web
- Hardware backends
- Uses connected model runtime
- Install methods
- Desktop application · Python package · Docker · Managed cloud
Models and integrations
- Model formats
- Uses connected model runtime
- Common integrations
- OllamaOpenAI-compatible APIsObsidianNotion
Strengths
- Knowledge, agents, and automation in one product
- Local and cloud model support
- Multiple personal productivity interfaces
Limitations
- Several integrations require configuration
- Self-hosting still needs model and storage planning
Planning checklist
Before you choose Khoj
Answer these questions with the exact models, hardware, users, and data you expect to operate.
- Does Khoj support the exact model, provider, data source, and operating system required by the workflow?
- Can the available hardware and memory handle the selected models, context, concurrency, and runtime overhead?
- Do the AGPL-3.0 terms fit internal use, modification, redistribution, and any commercial service being planned?
- Who will own upgrades, credentials, backups, monitoring, and recovery when this tool becomes part of a real workflow?
Khoj FAQ
What layer does Khoj replace?
Khoj primarily covers knowledge rag. It does not automatically replace every model runtime, application, storage service, or infrastructure dependency connected to that layer.
Can Khoj run entirely locally?
Yes, a local or self-hosted path is available. Privacy still depends on the model providers, connectors, telemetry, and external tools that you enable.
What should be tested before adopting Khoj?
Use the exact models, documents, integrations, hardware, concurrency, and security boundary expected in production. Feature lists and public benchmarks cannot validate that complete combination.
Official sources
Use these links to confirm current compatibility and installation requirements.
Related tools
Execution evidence
Known working recipes using Khoj
Recipes connect hardware, a model artifact, tools, settings, verification, and a reportable result.
No verified recipe is linked to this record yet.
Compatibility estimates remain available in the planner. A recipe appears here only after its exact stack and verification protocol are documented.