Langtrace AI website screenshot

Langtrace AI

Langtrace is an open-source, OpenTelemetry-based end-to-end observability platform for LLM applications, built by Scale3 Labs. It captures real-time traces, metrics, and evaluations for popular LLMs, agent frameworks, and vector databases. Traces are ingested via an OTLP/HTTP endpoint and a REST API exposes projects, prompt registry, and trace retrieval. Available as a free self-hosted deployment or as Langtrace Cloud.

4 APIs 0 Features
AILLMObservabilityOpenTelemetryTracingOpen Source

APIs

Langtrace Trace Ingestion (OTEL)

OpenTelemetry-compatible trace ingestion endpoint (POST /api/trace). Accepts JSON-encoded OTLP/HTTP spans authenticated with a project-level x-api-key header, for both Langtrace...

Langtrace Projects and API Keys

REST endpoints to create projects (POST /api/project) under a team and to mint project-level API keys (POST /api/api-key), authenticated with an x-api-key header.

Langtrace Prompt Registry

Fetch versioned prompts from the prompt registry (GET /api/promptset) with optional version selection and dynamic variable substitution, returning the live prompt when no versio...

Langtrace Trace Retrieval and Metrics

Paginated retrieval of stored traces for a project (POST /api/traces), returning span results plus pagination metadata. Underlying spans carry the token, cost, latency, and eval...

Collections

Pricing Plans

Langtrace Plans Pricing

4 plans

PLANS

Rate Limits

Langtrace Rate Limits

4 limits

RATE LIMITS

FinOps

Resources

👥
GitHubOrganization
GitHubOrganization
🔗
LinkedIn
LinkedIn
🔗
Website
Website
🔗
Documentation
Documentation
🔗
Plans
Plans
🔗
RateLimits
RateLimits
🔗
FinOps
FinOps

Sources

Raw ↑
opencollection: 1.0.0
info:
  name: Langtrace AI API
  version: '1.0'
request:
  auth:
    type: apikey
    key: x-api-key
    value: '{{apiKey}}'
    in: header
items:
- info:
    name: Traces
    type: folder
  items:
  - info:
      name: Send OpenTelemetry-compatible traces to Langtrace.
      type: http
    http:
      method: POST
      url: https://app.langtrace.ai/api/trace
      body:
        type: json
        data: '{"resourceSpans":[]}'
    docs: Ingests JSON-encoded OTLP/HTTP spans via an otlphttp exporter with JSON encoding and no compression. Authenticated
      with the x-api-key header.
  - info:
      name: Download traces for a project.
      type: http
    http:
      method: POST
      url: https://app.langtrace.ai/api/traces
      body:
        type: json
        data: '{"projectId":"<string>","page":1,"pageSize":100}'
    docs: Retrieves stored traces for a project, paginated. Up to 100 traces per page plus pagination metadata.
- info:
    name: Projects
    type: folder
  items:
  - info:
      name: Create a new project.
      type: http
    http:
      method: POST
      url: https://app.langtrace.ai/api/project
      body:
        type: json
        data: '{"teamId":"<string>","name":"<string>","description":"<string>","createDefaultTests":false}'
    docs: Creates a new project under the specified team.
  - info:
      name: Create an API key for an existing project.
      type: http
    http:
      method: POST
      url: https://app.langtrace.ai/api/api-key?project_id=<string>
      body:
        type: none
        data: ''
    docs: Mints a new project-level API key for the given project.
- info:
    name: Prompt Registry
    type: folder
  items:
  - info:
      name: Get a prompt from the registry.
      type: http
    http:
      method: GET
      url: https://app.langtrace.ai/api/promptset?promptset_id=<string>&version=1
    docs: Fetches a prompt set from the prompt registry. If no version is provided the live prompt is returned.