rouge-score
Pure python implementation of ROUGE-1.5.5.
Decision gist · record as of 2026-08-14
Yes, with conditions. rouge-score is the standard pure-Python ROUGE implementation and is actively maintained with no known vulnerabilities. Install it if you need ROUGE metrics for summarization evaluation and can tolerate high install friction from source-only distribution. The gap since the last release in 2022 is a minor concern for stability but not a blocker given active repository maintenance. Suitable for research, benchmarking, and production evaluation pipelines.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python >= 3.7; high install friction due to source-only distribution (no wheels).
- High install friction: the package is distributed as a source tarball with no wheels, requiring compilation or build tools on installation.
- Maintenance is active with recent commits, but the latest release was in 2022.
License · maintenance · safety
permissive license (permissive) — Licensed under Apache 2.0 (permissive), allowing commercial and private use with minimal restrictions. No notable licensing constraints for typical adoption.
last release 2022-07-22 (1484 days) · last repo commit 2026-08-13 · 38,531 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 3,486,792 downloads/mo, #2,605 on PyPI
Alternatives
Verify before relying
pip install rouge-score
from rouge_score import rouge_scorer
scorer = rouge_scorer.RougeScorer(['rouge1', 'rougeL'], use_stemmer=True)
scores = scorer.score('The quick brown fox jumps over the lazy dog',
'The quick brown dog jumps on the log.')- Whether the package's implementation continues to match the original perl ROUGE results given the time since last release.
- Performance characteristics when scoring large batches of summaries or very long texts.
- Whether bootstrap resampling for confidence intervals is documented with examples.
What it is and what it does
rouge-score is a pure Python implementation of ROUGE, the automatic evaluation metric for text summarization. It replicates the original Perl ROUGE package's behavior, implementing ROUGE-N (n-gram overlap), ROUGE-L (longest common subsequence at sentence level), and ROUGE-Lsum (summary-level LCS with union computation). The package also supports optional Porter stemming and bootstrap resampling for confidence intervals, with text normalization built in.
The package is designed for researchers and practitioners who need to evaluate generated summaries against reference summaries. It can be used programmatically via RougeScorer or as a command-line tool to batch-score target and prediction files. No external dependencies are required at runtime, though the source distribution has high install friction.
Use it for
- Evaluate machine-generated summaries in NLP research pipelines against gold-standard references.
- Benchmark summarization models during development and compare results with published papers.
- Batch-score large sets of predictions and targets from command line with CSV output.
- Compute ROUGE metrics with optional stemming to handle morphological variations in summary text.
- Calculate confidence intervals via bootstrap resampling to assess score stability.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, with conditions.
rouge-score is the standard pure-Python ROUGE implementation and is actively maintained with no known vulnerabilities. Install it if you need ROUGE metrics for summarization evaluation and can tolerate high install friction from source-only distribution. The gap since the last release in 2022 is a minor concern for stability but not a blocker given active repository maintenance. Suitable for research, benchmarking, and production evaluation pipelines.
Install
rouge-score on PyPI
Before you install
High install friction: the package is distributed as a source tarball with no wheels, requiring compilation or build tools on installation. Maintenance is active with recent commits, but the latest release was in 2022.
Requires Python >= 3.7; high install friction due to source-only distribution (no wheels).
License in practice
Licensed under Apache 2.0 (permissive), allowing commercial and private use with minimal restrictions. No notable licensing constraints for typical adoption.
Quickstart
pip install rouge-score
from rouge_score import rouge_scorer
scorer = rouge_scorer.RougeScorer(['rouge1', 'rougeL'], use_stemmer=True)
scores = scorer.score('The quick brown fox jumps over the lazy dog',
'The quick brown dog jumps on the log.')Verify before relying
- Whether the package's implementation continues to match the original perl ROUGE results given the time since last release.
- Performance characteristics when scoring large batches of summaries or very long texts.
- Whether bootstrap resampling for confidence intervals is documented with examples.
Package facts
| License | permissive license permissive |
| Python support | Supports the current Python release >=3.7 |
| Install friction | High. Source build required |
| Runtime dependencies | None |
| Maintenance | Actively maintained 1,484 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 3,486,792 / month, #2,605 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | License :: OSI Approved :: Apache Software LicenseOperating System :: OS IndependentProgramming Language :: Python :: 3 |
Evidence: rouge_score-0.1.2.tar.gz
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “n-gram overlap scoring”
- rouge-scoreComputes ROUGE scores (ROUGE-N, ROUGE-L, ROUGE-Lsum) to evaluate the…
- dtlpymetricsCalculates and generates quality scores for annotations in image and…
- strsimpyImplements a dozen string similarity and distance algorithms…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also rouge-metric · rouge · rouge-chinese · redlines · pycocoevalcap · bert-score · pylcs · rank-bm25 · strsimpy · Distance