TruLens
TruLens is a free, open-source Python SDK for tracing and evaluating AI agents, RAG applications, MCP apps, and summarization systems.
What TruLens is
TruLens is a free, open-source Python SDK for tracing and evaluating AI agents, RAG applications, MCP apps, and summarization systems. It records step-level latency, inputs, outputs, token use, and cost to help teams investigate poor answers. Evaluations include measures such as tool selection, groundedness, context relevance, and correctness. Users can add custom metrics with Python functions or LLM-as-judge prompts, and refine judges with rubrics and examples. Evaluations can run as traces arrive or in batches over datasets, while a leaderboard compares app versions on metrics, latency, and cost. TruLens operates on OpenTelemetry traces and can instrument custom Python apps as well as frameworks including LangGraph and LlamaIndex. Listed integrations and providers include LangChain, MCP, OpenAI, Anthropic, Amazon Bedrock, HuggingFace, LiteLLM, Google, and Snowflake Cortex. Data can be logged to local SQLite by default or to a SQLAlchemy-compatible database URL, and a Snowflake connector is available. The project was originally created by TruEra and is actively maintained as open source by Snowflake. It is self-hosted and available through pip or from source.
Good
- Records step latency, inputs, outputs, tokens, and cost.
- Evaluations support custom metrics and judge prompts.
- Can evaluate traces as they arrive or in batches.
- Logs to SQLite or a compatible database URL.
Watch out
- SnowflakeConnector lacks conversation grouping and metrics.
- SnowflakeConnector lacks retrieval APIs and dashboard views.
- Requires a self-hosted deployment.
Who it suits
TruLens suits teams building AI agents, RAG systems, MCP apps, or summarizers who need trace inspection and evaluation. It is aimed at Python-based workflows and can instrument custom apps and listed frameworks.
Verdict
TruLens combines tracing, evaluation, and version comparison in a self-hosted Python SDK. Teams using the Snowflake connector should note its missing conversation features, retrieval APIs, and dashboard views.
Plans and prices
From TruLens's own pricing page, checked 7 Oct 2026.
| Plan | Price | What it gives you |
|---|---|---|
| Open source | Free | Open source Python SDK · install with pip or from source |
| Open-source TruLens library | Free | Python library · self-hosted storage/dashboard · optional provider packages · No subscription fee; users supply their own model/API and infrastructure costs |
The facts
Each one read from TruLens's own pages, with the date it was checked.
What it does
- Acquisition
- Snowflake announced an agreement on May 22, 2024 to acquire TruEra's AI-observability platform.checked 15 Oct 2026
- Agent benchmark
- The homepage reports that Agent GPA caught 267 of 281 human-annotated errors, or 95%, on TRAIL and GAIA traces.checked 15 Oct 2026
- Agent metrics
- Built-in agent metrics include tool selection, tool calling, tool quality, plan adherence, plan quality, execution efficiency and logical consistency.checked 15 Oct 2026
- API documentation
- The project publishes a searchable API reference covering core classes, metrics, providers and connectors.checked 15 Oct 2026
- Batch evaluation
- Runs can replay a dataset and compute the same metrics over stored inputs.checked 15 Oct 2026
- Current stewardship
- The official TruLens homepage says the project was created by TruEra and is actively maintained in open source by Snowflake after the acquisition.checked 15 Oct 2026
- Custom evaluation
- Developers can define metrics with Python functions, LLM-as-judge prompts, span attributes, rubrics and examples.checked 15 Oct 2026
- Custom metrics
- Users can define custom metrics with Python functions or LLM-as-judge prompts and tune judges with rubrics and examples.checked 7 Oct 2026
- Dashboard
- The built-in Streamlit dashboard provides leaderboards, trends, individual records and side-by-side version comparison.checked 15 Oct 2026
- Database connector
- A Snowflake connector is provided for storing TruLens traces and evaluation data.checked 15 Oct 2026
- Evaluation
- It supports agent, RAG, MCP app, and summarization evaluations with metrics including tool selection, groundedness, context relevance, and correctness.checked 7 Oct 2026
- Evaluation modes
- Evaluations can run as traces arrive or in batch over a dataset.checked 7 Oct 2026
- Experiment comparison
- The leaderboard compares app versions across metrics, latency, and cost.checked 7 Oct 2026
- Explained scores
- Its judges attach explanations to metric scores so teams can investigate why a result was rated poorly.checked 15 Oct 2026
- Founders
- Snowflake's announcement names Anupam Datta, Shayak Sen and Will Uppington as TruEra's three co-founders.checked 15 Oct 2026
- Framework integrations
- Official integrations instrument LangChain, LangGraph, LlamaIndex and NeMo Guardrails applications.checked 15 Oct 2026
- Installation
- The documented installation uses Conda with Python 3, followed by pip install trulens; source development uses the GitHub repository and editable installation.checked 15 Oct 2026
- Integrations
- The documentation lists integrations and providers for LangChain, LangGraph, LlamaIndex, MCP, OpenAI, Anthropic, Amazon Bedrock, HuggingFace, LiteLLM, Google, and Snowflake Cortex.checked 7 Oct 2026
- Licence
- The public GitHub repository identifies TruLens as MIT-licensed.checked 15 Oct 2026
- Limit: feature depth and scale
- Conversion Gems characterizes the product as having fewer features than some alternatives, less depth and possible slowness at scale.checked 15 Oct 2026
The company
- Company facts
- LinkedIn lists TruEra as a privately held software-development company founded in 2019 with headquarters in Redwood City, California, and 51–200 employees.checked 15 Oct 2026
- Independent company profile
- Craft describes TruEra as an AI-observability company founded in 2019 and headquartered in Redwood City.checked 15 Oct 2026
- Limit: operational ownership
- The independent review says production users must manage database lifecycle, access control, retention, upgrades and evaluator-worker reliability.checked 15 Oct 2026
Money
- Limit: external evaluation costs
- Evaluation can add model-token, infrastructure, database-retention and Snowflake costs, depending on deployment choices.checked 15 Oct 2026
Alternatives to TruLens
- 1
DeepEval
Free planOpen sourceWebAPILinuxmacOS8.3From $200 a month - 2
Promptfoo
Free planOpen sourceWebAPILinuxmacOS7.8Free plan - 3
Braintrust
Free planWebAPISelf-hosted7.7From $249 a month - 4
Giskard
Free planOpen sourceWebAPILinuxSelf-hosted7.6Free plan - 5
Arena (formerly Chatbot Arena)
Free planOpen sourceWebAPISelf-hosted7.1Free plan - 6
Maxim AI
Free planWebAPISelf-hosted7.0From $29 a month