Skip to content
AgentsIndex

TruLens

Open-source evaluation for AI agents with trace-level scoring.

GitHub stars
3.6k

What we checked

Pages last read
Facts checked
2 facts

From 2 sources

AI answers
Too few read

58 buyer-question answers read, November 2026

At a glance

Tags
Open source

Code

Last push
12 October 2026

For developers

Works with
OpenTelemetry

What is TruLens?

TruLens is an open-source AI evaluation tool for teams that measures agent quality by tracing execution flow and scoring retrieved context, tool calls, plans, and outputs. Its Evaluate, Iterate, and Test workflow helps teams compare versions and catch regressions with groundedness, context relevance, coherence, and answer relevance. It works through the Python SDK or OpenTelemetry traces and is used by teams at Equinix, Snowflake, and KBC Group.

What it is and who maintains it

TruLens instruments an AI agent or RAG pipeline with a Python decorator, records every span (latency, inputs, outputs, tokens, cost) as OpenTelemetry data, and scores steps with LLM-as-judge metrics. It was created by TruEra, which open-sourced it in 2021; Snowflake acquired TruEra in May 2024 and now stewards the project (snowflake.com/en/blog/snowflake-acquires-truera-to-bring-llm-ml-observability-to-data-cloud). It remains MIT-licensed and free — there is no separate paid tier of TruLens itself.

Tracing architecture: OpenTelemetry-native

Every instrumented call becomes a structured OTEL span, so traces are portable to any OTLP-compatible backend (Jaeger, Grafana Tempo, Datadog) rather than being locked in a proprietary format. This is a genuine differentiator among LLM-eval libraries, most of which use a custom trace schema. TruLens 2.9.0 (23 July 2026) added dual-emission of standard gen_ai.* OTel GenAI semantic-convention attributes, and 2.10.0 (28 July 2026) corrected span-kind values to match that spec, improving interoperability with OTel-native tooling (github.com/truera/trulens/releases/tag/trulens-2.10.0).

Judge quality: one independent benchmark, one vendor-authored one

AIMultiple, an independent research firm, benchmarked five context-relevance evaluators (TruLens, WandB Weave, Ragas, DeepEval, UpTrain) across 1,010 adversarial QA samples in March 2026. TruLens led on ranking quality — NDCG@5 of 0.932, Spearman ρ of 0.750, and the highest discrimination ratio (4.2:1) between a correct passage and a near-identical entity-swapped fake — while WandB Weave had the highest raw Top-1 accuracy due to its coarser binary scoring (aimultiple.com/rag-evaluation-tools). Separately, TruLens's own site cites a 95%-error-coverage figure on the TRAIL/GAIA agent benchmark from 'Agent GPA', a framework described in an arXiv paper authored by the TruEra/Snowflake research team, including TruEra co-founder Anupam Datta (arxiv.org/abs/2510.08847) — that number is the vendor's own research, not third-party corroboration, though it is publicly posted and citable.

Adoption evidence and how it's sourced

TruLens's GitHub repo maintains an ADOPTERS.md file that separates 'published references' (companies with a public blog post or press quote, e.g. Walmart Global Tech's RAG-triad case study, Thomson Reuters, phData, the SOA Research Institute) from 'community-reported usage' sourced from named engineers' own GitHub issues (Cisco, J.P. Morgan Chase, VMware by Broadcom, Hitachi Digital Services, and others) — and lists three logos it removed for lack of evidence, including Snowflake itself, explicitly because Snowflake maintains the project and isn't an independent adopter (github.com/truera/trulens/blob/main/ADOPTERS.md). That level of self-imposed sourcing discipline on a vendor's own logo wall is unusual and worth taking at face value; most of the individual claims (Equinix's '2 weeks to 2 hours' iteration-time reduction, KBC Group's early-adopter quote) trace to a March 2024 press release rather than an audited study, so they remain attributed vendor/customer claims, not verified metrics.

Release pace and maintenance

The repo is actively maintained: five tagged releases between 23 July 2026 and 20 August 2026 (2.9.0 through 2.13.1), roughly weekly, each shipping features from a rotating set of contributors — 2.9.0 and 2.10.0 each list 8-14 first-time contributors credited with shipping the release's headline features (github.com/truera/trulens/releases). Current PyPI release is 2.13.1, published 20 August 2026 (pypi.org/pypi/trulens/json).

Cost and what you're actually installing

TruLens itself is free and MIT-licensed; pip install trulens costs nothing (pypi.org/project/trulens). What you pay for is infrastructure you already run (a place to store traces) plus the LLM API calls the judges make when scoring your traces — TruLens 2.11.0 (4 August 2026) added a configure_online_eval() sampling/throttle/cost-budget mechanism specifically because always-on judging can be expensive on high-traffic apps (github.com/truera/trulens/releases/tag/trulens-2.11.0). Snowflake also offers a Cortex AI Observability feature inside its own platform; whether that is a separately-priced managed version of TruLens or a distinct product, and what it costs, is not established from public sources.

Where it doesn't fit

TruLens is a library you instrument yourself and either self-host the dashboard for or query programmatically — there is no vendor-hosted TruLens SaaS with a signup page, so teams wanting a zero-ops managed eval platform will need to look elsewhere or go through Snowflake's own product. Its PyPI package still carries a 'Development Status :: 3 - Alpha' trove classifier despite five years of history and version 2.13.1 (pypi.org/pypi/trulens/json) — worth noting as a calibration point even though the release cadence and adopter evidence suggest the classifier is stale rather than a live warning.

TruLens price and terms

Terms

Free plan
YesYesTruLens is an open-source library published under the MIT license
The terms in full

the site offers no paid plan, no pricing page and no usage limits. Model costs for LLM judges are paid to whichever provider the user configures.

Where these prices come from

trulens.org, checked

All plans and limits

PlanPriceMain limits
TruLensFreeNone stated

Read from trulens.org.

What we checked on TruLens

YesPartlyNo

What it costs to startYes

Free to use

TruLens is an MIT-licensed open-source library with no paid plan anywhere on its site; using it costs nothing beyond the LLM provider fees for the judges you run.

Open source, OpenTelemetry-native.

trulens.orgChecked

Questions about TruLens

How much does TruLens cost?
TruLens is free to use. From trulens.org, checked 18 November 2026.
Does TruLens have an API or an MCP server?
TruLens has an API.
Is TruLens open source, and can you host it yourself?
TruLens is open source. We checked github.com/truera/trulens/blob/main/LICENSE ourselves on 15 November 2026.
What have we checked on TruLens?
2 facts, under these headings: What it costs to start, how you build with it.
How do we check TruLens?
We hold 2 facts on TruLens, quoted from 2 pages. Facts and price last checked 18 November 2026, when we also found the last change.

How we check every listing

Sources and changes

Last changed

Sources1

What changed2

  • What it does
  • What it does

Repository

Licence
MIT
Stars
3,598
Last push
12 October 2026

TruLens price history

We have read the price once so far, on 18 November 2026: free.

PricePlanFirst seenLast seenReadingsSource
Free18 November 202618 November 20261Pricing page, read 18 November 2026

From our research on 18 November 2026. A date is when we read the page, not when the price changed.

Similar in Agent tools and integrations

See all in Agent tools and integrations

From our research

All research

Is TruLens yours?

Free, and the page then carries the verified owner mark. You confirm it with an email at trulens.org. Claim this listing

Report an error if a fact is wrong.