TruLens

license

MIT(OSI)

source

public(github.com/truera/trulens)

commercial

Snowflake-Cortex-AI-Observability(separate hosted product)

library

not-feature-gated

core features withheld

no

MIT library that is fully functional standalone; the Snowflake Cortex hosted product is separate, not a gate on the library.

• https://github.com/truera/trulens recorded 2026-08-13

  Repository page. The license sidebar reads "MIT license" and GitHub's own metadata records license spdxId "MIT", name "MIT License", resolved from the single root LICENSE file. The README still leads on feedback functions and the RAG Triad. 3,508 stars.

• https://pypi.org/project/trulens/ recorded 2026-09-29

  PyPI project page for trulens: license sidebar reads "MIT License (MIT)" and the classifier list includes "OSI Approved :: MIT License"; searched for enterprise, paid tier, pricing and license key: none.

• https://www.trulens.org/ recorded 2026-09-29

  Homepage tagline: "Open source, OpenTelemetry-native, MIT licensed." and "Since the acquisition, Snowflake actively maintains TruLens" in open source; searched the body for enterprise, license key, pricing and paid tier: none.

Adoption is measured by trulens' PyPI downloads in the trailing 30 days. Nothing else in the recorded evidence - no star count, customer roster or other channel - supports a higher reading.

• https://pypistats.org/api/packages/trulens/recent recorded 2026-10-07

  20,121 downloads in the trailing 30 days for trulens (pypistats last_month)

Evaluation-first observability: judge-scored feedback functions for RAG, agents, MCP apps and summarization, custom metrics, OpenTelemetry tracing and side-by-side comparison of app versions. It is narrower than Langfuse as a platform, with no prompt management or annotation queues in the docs, but its evaluation depth makes up the difference, so it sits level with it.

• https://github.com/truera/trulens recorded 2026-08-13

  README - feedback functions built around "the RAG Triad" (groundedness, context relevance, answer relevance) with links to Snowflake's RAG-triad benchmarks, and a comparison claiming TruLens leads on three of four ranking metrics ahead of WandB Weave, RAGAS, DeepEval and UpTrain. No AI gateway, prompt registry or annotation-queue feature appears.

• https://www.trulens.org/ recorded 2026-09-29

  Homepage of the TruLens docs site. It describes TruLens as an evaluation and tracing tool, saying "Judges explain the score, so a low score" points to what went wrong, records "Latency, inputs, outputs, tokens and cost, recorded per step", and offers custom metrics from "Any Python function" and "Any LLM-as-judge prompt". The RAG use case asks "Is the answer supported by what was retrieved?", and the version section is headed "Pick a winner" with a leaderboard.

• https://www.trulens.org/getting_started/ recorded 2026-09-29

  Getting Started docs page, navigation. It links quickstarts titled "Ground Truth Evaluations" and "Blocking Guardrails Quickstart", provider pages for "Snowflake Cortex" and "LiteLLM", and a dashboard reference at "../reference/trulens/dashboard/streamlit/".