AI Potluck
Product / UX / Telemetry & observability

Langtrace

Scale3 Labs

OpenTelemetry-based end-to-end observability tool for LLM applications, providing real-time tracing, evaluations and metrics across model providers, frameworks and vector databases, with TypeScript and Python SDKs. It is distinguished by strict OTel compliance and vector-database tracing, and deploys from the repository's docker-compose file.

The application and the SDKs are distributed under different terms; the openness axis follows the application's. Verified 2026-08-13 via the Scale3-Labs/langtrace repository.

Openness

5 high confidence
5.0
license
AGPL-3.0(OSI, copyleft, the application)
source
public
core-gated
ungated(the whole AGPL-3.0 application is in the repository - app, components, lib, prisma, proto, Dockerfile, docker-compose.yaml - with no ee/ or enterprise directory, no second license and no license-key check, and the vendor's FAQ says Langtrace Cloud is not charged for)

An AGPL-3.0 application with Apache-2.0 SDKs, and nothing withheld from the published source, which is the top rung: an OSI license, public source, no gated core. The whole application is in the repository, with no enterprise directory, no second license and no license key; self-hosting is documented as a first-class path; and the vendor's own FAQ says of the hosted tier "Currently, we are not charging anything for Langtrace cloud". Neither argument for holding it lower survives: AGPL copyleft sits on the OSI rung alongside Apache, and a hosted product sold beside an otherwise complete open core does not gate that core.

  • https://github.com/Scale3-Labs/langtrace recorded 2026-06-04

    AGPL-3.0 app license, Apache-2.0 SDKs, OTel-based observability

  • https://github.com/Scale3-Labs/langtrace/releases recorded 2026-06-04

    app v4.0.11 release 17 Apr 2026 (active)

  • https://api.github.com/repos/Scale3-Labs/langtrace/contents/ recorded 2026-08-12

    Repository root, listed to test for a carve-out. Dockerfile, LICENSE, README.md, SECURITY.md, app, components, lib, prisma, proto, public, scripts, middleware.ts, docker-compose.yaml and the Next.js config files. There is no ee/, no enterprise/, no second LICENSE below the root, and no license-key module. GitHub's license endpoint returns AGPL-3.0 for the single root LICENSE. The published tree is the whole application and it self-hosts from the docker-compose file in it.

  • https://raw.githubusercontent.com/Scale3-Labs/langtrace/main/README.md recorded 2026-08-12

    The vendor's own statement of the split and of the cloud tier. License section: "Langtrace application(this repository) is licensed under the AGPL 3.0 License" and "Langtrace SDKs are licensed under the Apache 2.0 License." The feature list includes "Self-hosting Option: Deploy on your own infrastructure", and the FAQ answers "Can I self host and run Langtrace in my own cloud?" affirmatively. On price: "What is the pricing for Langtrace cloud? Currently, we are not charging anything for Langtrace cloud and we are primarily looking for feedback so we can continue to improve the project. We will inform our users when we decide to monetize it." A cloud tier that charges nothing cannot be the paid tier a core is withheld for.

Adoption

2 medium confidence
2.0

langtrace-python-sdk draws roughly 34,500 PyPI downloads a month, inside the 10K-100K band, and the project has around 1,200 GitHub stars. A real but small footprint, at early-adopter tier.

Capability

3 medium confidence
3.0

Strong on OpenTelemetry-native real-time tracing across LLMs, frameworks and vector databases, with evaluations, latency, cost and usage analytics and broad integrations (OpenAI, Anthropic and Cohere; LangChain and LlamaIndex; Pinecone, Chroma and Weaviate). Prompt versioning and dataset or annotation workflows are thinner or unclear, which leaves it one rung below Langfuse at 4.

  • https://github.com/Scale3-Labs/langtrace recorded 2026-08-13

    Repository description - "observability tool for LLM applications, providing real-time tracing, evaluations and metrics for popular LLMs, LLM frameworks, vectorDBs and more" - with repository topics including datasets, evaluations, llmops and observability. The README Features list itself names Open Telemetry Support, Real-time Monitoring, Performance Insights and Debug Tools.

Verified 2026-08-12