Klu website screenshot

Klu

Klu (klu.ai) is an LLM app platform for designing, deploying, evaluating, and observing prompt-driven AI applications. The Klu Engine exposes a REST API at https://api.klu.ai/v1 (Bearer API key) where an Action encapsulates a prompt template, model config, context (RAG), and output parsing, and is invoked to generate completions, with data, feedback, sessions, and models managed alongside it. NOTE - product development appears dormant - the official klu SDK last shipped in March 2025 and the founding team's recent open-source work has shifted toward xAI / Grok tooling - see review.yml for the honest status assessment.

5 APIs 0 Features
AILLMLLM App PlatformPrompt EngineeringEvaluationObservability

APIs

Klu Actions / Prompt API

Runs a Klu Action by GUID against input variables, returning a generated completion plus a feedback URL. Supports streaming, async execution, response caching, session memory, c...

Klu Data / Feedback API

Captures and manages the data points generated by Actions - each holding the prompt, completion, and metadata - and the feedback (ratings, corrections, issues) attached to them....

Klu Models API

Manages the LLM providers and models available in a workspace and the default model used by Actions. Klu brokers across 50+ models and providers (OpenAI, Anthropic, and others) ...

Klu Sessions API

Stores and retrieves session memory across multiple Action generations, enabling multi-turn conversational experiences such as copilots and coaches that retain context between r...

Klu Apps / Workspaces API

Organizes work into Workspaces and Apps (projects) that group related Actions, context, and experiments. Apps are the top-level container for an AI use case; experiments run pro...

Collections

Klu API

OPEN

Pricing Plans

Klu Ai Plans Pricing

3 plans

PLANS

Rate Limits

Klu Ai Rate Limits

3 limits

RATE LIMITS

FinOps

Klu Ai Finops

FINOPS

Resources

🔗
DomainSecurity
DomainSecurity
🔑
Authentication
Authentication
👥
GitHubOrganization
GitHubOrganization
🔗
LinkedIn
LinkedIn
🔗
Website
Website
🔗
Documentation
Documentation
🔗
Plans
Plans
🔗
RateLimits
RateLimits
🔗
FinOps
FinOps

Sources

Raw ↑
opencollection: 1.0.0
info:
  name: Klu API
  version: '1.0'
request:
  auth:
    type: bearer
    token: '{{apiKey}}'
items:
- info:
    name: Actions
    type: folder
  items:
  - info:
      name: Run an Action
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/actions
      body:
        type: json
        data: "{\n  \"action\": \"<action_guid>\",\n  \"input\": \"<string or key-value object>\",\n  \"streaming\": false,\n\
          \  \"async_mode\": false,\n  \"cache\": false\n}"
    docs: Executes a Klu Action by GUID against the provided input, returning the generated completion and a feedback URL.
- info:
    name: Context
    type: folder
  items:
  - info:
      name: Create a Context library
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/context
      body:
        type: json
        data: "{\n  \"name\": \"<string>\",\n  \"description\": \"<string>\",\n  \"type\": \"Custom\"\n}"
    docs: Creates a new Context library for retrieval-augmented generation.
  - info:
      name: Add a document to a Context library
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/context/{guid}/documents
      body:
        type: json
        data: "{\n  \"text\": \"<string>\",\n  \"meta_data\": {}\n}"
    docs: Adds app data as a Context document with optional metadata.
  - info:
      name: Add files to a Context library
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/context/{guid}/add_files
      body:
        type: json
        data: "{\n  \"files\": []\n}"
    docs: Adds one or more files to an existing Context library.
- info:
    name: Data
    type: folder
  items:
  - info:
      name: Get a data point
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/data/{guid}
    docs: Retrieves a data point by GUID, including its prompt, completion, and metadata.
- info:
    name: Feedback
    type: folder
  items:
  - info:
      name: Submit feedback on a data point
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/feedback
      body:
        type: json
        data: "{\n  \"data_guid\": \"<data_point_guid>\",\n  \"type\": \"rating\",\n  \"value\": 2\n}"
    docs: Submits feedback for a generation. Ratings use 1 (negative) and 2 (positive).
- info:
    name: Sessions
    type: folder
  items:
  - info:
      name: Create a session
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/sessions
      body:
        type: json
        data: "{\n  \"action\": \"<action_guid>\",\n  \"name\": \"<string>\"\n}"
    docs: Creates a session used to store and retrieve memory across Action generations.
  - info:
      name: Get a session
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/sessions/{guid}
    docs: Retrieves a session by GUID.
- info:
    name: Models
    type: folder
  items:
  - info:
      name: List models
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/models
    docs: Lists the LLM providers and models available in the workspace.
  - info:
      name: Get a model
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/models/{guid}
    docs: Retrieves a model by GUID.
- info:
    name: Apps
    type: folder
  items:
  - info:
      name: List apps
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/apps
    docs: Lists the Apps (projects) in the workspace.
  - info:
      name: Create an app
      type: http
    http:
      method: POST
      url: https://api.klu.ai/v1/apps
      body:
        type: json
        data: "{\n  \"name\": \"<string>\",\n  \"description\": \"<string>\",\n  \"app_type\": \"<string>\"\n}"
    docs: Creates an App (project).
  - info:
      name: Get an app
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/apps/{guid}
    docs: Retrieves an App by GUID.
- info:
    name: Workspaces
    type: folder
  items:
  - info:
      name: List workspaces
      type: http
    http:
      method: GET
      url: https://api.klu.ai/v1/workspaces
    docs: Lists the workspaces accessible to the API key.