Humanloop

aid: humanloop
name: Humanloop
description: >-
  Humanloop was an LLM development platform for managing prompts, datasets,
  evaluations, and production observability of LLM applications, used by AI
  product teams at companies like Duolingo and Gusto. The platform offered a
  collaborative prompt editor, versioned prompts and tools, evaluation runs
  (LLM-as-judge and code-based), dataset management, online evaluators, and
  full request logging across OpenAI, Anthropic, Google, and other providers,
  with Python and TypeScript SDKs and a REST API. In 2025 the Humanloop team
  was acquired by Anthropic and the company has announced it is sunsetting
  the standalone platform; an official migration guide is published for
  existing customers. This record captures the historical API surface for
  archival and discovery purposes.
type: Index
position: Provider
access: 3rd-Party
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
  - LLM Platform
  - Prompt Management
  - Evaluations
  - LLM Ops
  - Observability
  - Datasets
  - Prompts
  - Tools
  - Logs
  - Multi-Provider
  - Sunsetting
  - Acquired
url: https://raw.githubusercontent.com/api-evangelist/humanloop/refs/heads/main/apis.yml
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.20'
apis:
  - aid: humanloop:llm-platform
    name: Humanloop LLM Platform API
    description: >-
      The Humanloop REST API and SDKs covered prompts, tools, datasets,
      evaluations, evaluators, and logs for LLM applications. Developers
      called prompt endpoints from production to capture logs, ran
      evaluations against datasets, and managed versioned prompt
      configurations. The platform is being sunset following the team's
      acquisition by Anthropic in 2025.
    humanURL: https://humanloop.com/docs
    baseURL: https://api.humanloop.com/v5
    tags:
      - Prompts
      - Tools
      - Datasets
      - Evaluations
      - Evaluators
      - Logs
      - Files
    properties:
      - type: Documentation
        url: https://humanloop.com/docs
      - type: APIReference
        url: https://humanloop.com/docs/api-reference
      - type: SDK
        url: https://github.com/humanloop/humanloop-python
      - type: SDK
        url: https://github.com/humanloop/humanloop-typescript
      - type: GitHubOrganization
        url: https://github.com/humanloop
      - type: MigrationGuide
        url: https://humanloop.com/docs/guides/migrating-from-humanloop
      - type: SunsetAnnouncement
        url: https://humanloop.com
    features:
      - name: Prompt Management
        description: Versioned, collaborative prompt editor with structured variables and side-by-side comparisons.
      - name: Evaluations
        description: Run dataset-driven evaluations with LLM-as-judge and code-based evaluators.
      - name: Datasets
        description: Manage curated examples and production samples for evaluations and fine-tuning.
      - name: Logs and Observability
        description: Capture every LLM call with inputs, outputs, costs, and latencies for production debugging.
      - name: Online Evaluators
        description: Score live traffic in production with custom evaluators.
      - name: Tools and Functions
        description: Define reusable tools and function specs that prompts can call.
      - name: Multi-Provider Routing
        description: Use OpenAI, Anthropic, Google, and other providers through a single API surface.
      - name: Python and TypeScript SDKs
        description: First-class SDKs for invoking prompts and logging from application code.
    useCases:
      - name: Production Prompt Iteration
        description: Update prompts in Humanloop without redeploying application code.
      - name: LLM Evaluation
        description: Score prompts and models against curated datasets before shipping.
      - name: Production Monitoring
        description: Log every LLM call and trace quality regressions in production.
      - name: Cross-Functional Collaboration
        description: Let PMs, designers, and engineers collaborate on prompts in one workbench.
    integrations:
      - name: OpenAI
      - name: Anthropic
      - name: Google
      - name: Cohere
      - name: AWS Bedrock
      - name: Azure OpenAI
      - name: LangChain
      - name: LlamaIndex
    authentication:
      - type: API Key
        description: Workspace API keys passed via the `X-API-KEY` header authenticated REST and SDK calls.
common:
  - type: Website
    url: https://humanloop.com
  - type: Documentation
    url: https://humanloop.com/docs
  - type: Blog
    url: https://humanloop.com/blog
  - type: GitHubOrganization
    url: https://github.com/humanloop
  - type: MigrationGuide
    url: https://humanloop.com/docs/guides/migrating-from-humanloop
  - type: MediaKit
    url: https://humanloop.com/media-kit
  - type: TermsOfService
    url: https://humanloop.com/terms
  - type: PrivacyPolicy
    url: https://humanloop.com/privacy
  - type: Twitter
    url: https://x.com/humanloop
  - type: LinkedIn
    url: https://www.linkedin.com/company/humanloop
maintainers:
  - FN: Kin Lane
    email: [email protected]