Ragas

license

Apache-2.0(OSI)

source

public(full metrics + test-gen library)

core features withheld

no

Ragas is Apache-2.0 and fully public. The README installs the whole metric suite from PyPI or a git checkout, and the setup needs no account or license key.

• https://api.github.com/repos/vibrantlabsai/ragas/license recorded 2026-08-13

  GitHub's license endpoint reports spdx_id Apache-2.0 for vibrantlabsai/ragas and returns the Apache-2.0 text as the LICENSE body.

• https://github.com/explodinggradients/ragas recorded 2026-06-04

  Apache-2.0 license; LLM-app eval metrics + test-set generation; v0.4.3 (Jan 2026); 14.2k stars

• https://raw.githubusercontent.com/vibrantlabsai/ragas/main/LICENSE recorded 2026-09-29

  LICENSE file is the Apache License text: "Apache License" heading followed by "Version 2.0, January 2004", with the copyright line "Copyright [2023] [Vibrant Labs]".

• https://raw.githubusercontent.com/vibrantlabsai/ragas/main/pyproject.toml recorded 2026-09-29

  pyproject.toml describes the package as "Evaluation framework for RAG and LLM applications", points its license at the repository LICENSE file, and lists open-source packages (numpy, datasets, pydantic, openai, langchain) under the core "# Core dependencies" block, with no proprietary or license-key package.

• https://raw.githubusercontent.com/vibrantlabsai/ragas/main/README.md recorded 2026-09-29

  README installs the package with "pip install ragas" or "pip install git+https://github.com/vibrantlabsai/ragas" and shows the metric classes "DiscreteMetric" used directly after that install, with no Ragas account, Ragas key or paid tier in the setup (the only key mentioned is the user's own OPENAI_API_KEY for the judge model); README carries the "License Apache-2.0" badge.

ragas' PyPI downloads are the primary adoption signal, and the tool functions as the de facto standard for RAG and LLM-application evaluation, widely integrated with LangChain, LlamaIndex and observability platforms. GitHub stars, now under the project's new vibrantlabsai organization, corroborate that reading.

• https://github.com/explodinggradients/ragas recorded 2026-06-04

  14.2k stars; LangChain/observability integrations

• https://pypistats.org/api/packages/ragas/recent recorded 2026-08-13

  last_month = 1,587,521 downloads of the ragas package (last_week 335,773, last_day 58,054)

• https://pypistats.org/packages/ragas recorded 2026-06-04

  1,513,553 downloads in the last month

Ragas ships about thirty metrics grouped by use, including retrieval-augmented generation, agents and tool use, SQL and general-purpose metrics, plus automatic test-set generation and support for writing custom metrics. It is scoped to LLM applications and RAG pipelines, so it covers less benchmark ground than HELM.

• https://github.com/explodinggradients/ragas recorded 2026-06-04

  objective metrics (LLM-based + traditional), Aspect Critique, automatic test-data generation, integrations

• https://raw.githubusercontent.com/vibrantlabsai/ragas/main/docs/concepts/metrics/available_metrics/index.md recorded 2026-09-29

  Available-metrics page lists the shipped metrics in groups (Retrieval Augmented Generation, Nvidia Metrics, Agents or Tool use cases, Natural Language Comparison, SQL, General purpose) and says "One can also modify or write your own metrics using ragas."; General purpose lists "Aspect critic" and the RAG group lists "Faithfulness" and "Context Precision".

• https://raw.githubusercontent.com/vibrantlabsai/ragas/main/README.md recorded 2026-09-29

  README leads with "Objective metrics, intelligent test generation, and data-driven insights for LLM apps", lists "Objective Metrics: Evaluate your LLM applications with precision using both LLM-based and traditional metrics" and "Test Data Generation: Automatically create comprehensive test datasets covering a wide range of scenarios", installs with "pip install ragas" or from the git repo, and demonstrates Aspect Critique through "DiscreteMetric".