Skip to content

Add LiteLLM as AI gateway #1315

@RheagalFire

Description

@RheagalFire

Describe the feature

Add LiteLLM as a new chat model provider type alongside OpenAI, Anthropic, and Google. LiteLLM provides a unified completion() API that routes to 100+ LLM providers (AWS Bedrock, Azure, Ollama, Mistral, Cohere, Groq, DeepSeek, etc.) using the same OpenAI-compatible interface. This would let self-hosted Khoj users access any LLM provider without needing individual provider implementations.

Use Case

  • Self-hosted users currently limited to OpenAI, Anthropic, and Google. Adding each new provider requires code changes.
  • Users hitting issues with OpenAI-compatible proxies. LiteLLM handles provider-specific quirks internally.
  • Users wanting to use local models via Ollama, or enterprise models via AWS Bedrock/Azure, without configuring them as OpenAI-compatible endpoints.

Proposed Solution

No response

Additional Information

No response

Link to Discord or Github discussion

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions