LangSmith

aid: langsmith
url: https://raw.githubusercontent.com/api-evangelist/langsmith/refs/heads/main/apis.yml
name: LangSmith
x-type: company
description: >-
  LangSmith is the observability, debugging, and evaluation platform for LLM applications, built by LangChain. The LangSmith API exposes tracing, dataset management, evaluation, prompt-hub, and Fleet agent functionality for AI engineering teams.
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
  - AI
  - LLM
  - Observability
  - Evaluations
  - LangChain
created: '2026-05-08'
modified: '2026-05-08'
specificationVersion: '0.19'
apis:
  - aid: langsmith:langsmith-tracing-api
    name: LangSmith Tracing API
    tags:
      - Tracing
      - Observability
      - LLM
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/tracing
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith
        type: Documentation
      - url: https://docs.langchain.com/langsmith/tracing
        type: API Reference
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Capture, ingest, and inspect traces for LLM, agent, and chain executions. Traces include nested runs (spans), latency, token counts, errors, inputs/outputs, and metadata. Tracing is the primary unit of pricing on LangSmith.
  - aid: langsmith:langsmith-datasets-api
    name: LangSmith Datasets API
    tags:
      - Datasets
      - Examples
      - Evaluations
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/datasets
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith/datasets
        type: Documentation
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Manage datasets and example records used as ground-truth or test inputs for evaluating LLM applications. Supports CRUD on datasets, examples, and dataset splits.
  - aid: langsmith:langsmith-evaluations-api
    name: LangSmith Evaluations API
    tags:
      - Evaluations
      - Experiments
      - LLM
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/evaluation
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith/evaluation
        type: Documentation
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Run offline and online evaluations against datasets, attach feedback and scores to runs, and compare experiments. Supports LLM-as-judge, code-based, and human evaluators.
  - aid: langsmith:langsmith-prompts-api
    name: LangSmith Prompt Hub API
    tags:
      - Prompts
      - Prompt Management
      - LLM
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/prompt-hub
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith/prompt-hub
        type: Documentation
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Versioned prompt repository (Prompt Hub) for storing, retrieving, and collaborating on LLM prompts. Supports tagged versions, public/private prompts, and pull/push from SDKs.
  - aid: langsmith:langsmith-feedback-api
    name: LangSmith Feedback API
    tags:
      - Feedback
      - Scoring
      - Evaluations
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/feedback
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith/feedback
        type: Documentation
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Attach human or programmatic feedback (scores, comments, correction labels) to runs and trace nodes for evaluation, monitoring, and reinforcement signal collection.
  - aid: langsmith:langsmith-annotation-queues-api
    name: LangSmith Annotation Queues API
    tags:
      - Annotation
      - Human Review
      - Evaluations
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/annotation-queues
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith/annotation-queues
        type: Documentation
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Route runs to human reviewers via annotation queues. Reviewers grade outputs, attach corrections, and feed labels back into datasets for evaluation and fine-tuning.
  - aid: langsmith:langsmith-fleet-api
    name: LangSmith Fleet (Agent Deployment) API
    tags:
      - Agents
      - Deployment
      - LangGraph
    image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
    humanURL: https://docs.langchain.com/langsmith/deployments
    baseURL: https://api.smith.langchain.com
    properties:
      - url: https://docs.langchain.com/langsmith/deployments
        type: Documentation
      - url: openapi/langsmith-openapi.json
        type: OpenAPI
    description: >-
      Deploy and manage LangGraph agents in production via Fleet. Provides agent invocation, run management, scheduled jobs, and uptime billing for hosted agent deployments.
common:
  - type: Website
    url: https://smith.langchain.com/
  - type: Documentation
    url: https://docs.langchain.com/langsmith
  - type: Pricing
    url: https://www.langchain.com/pricing
  - type: Plans
    url: plans/langsmith-plans-pricing.yml
  - type: RateLimits
    url: rate-limits/langsmith-rate-limits.yml
  - type: FinOps
    url: finops/langsmith-finops.yml
maintainers:
  - FN: Kin Lane
    email: [email protected]