Open-rewrite

OpenRewriteEval is a benchmark for evaluating open-ended rewriting of long-form texts, covering a wide variety of rewriting types expressed through natural language instructions including formality, expansion, conciseness, paraphrasing, and tone and style transfer.

Llama 3.2 3B Instruct from Meta currently leads the Open-rewrite leaderboard with a score of 0.401 across 1 evaluated AI models.

Paper
About this benchmark

What Open-rewrite measures

Open-rewrite is a text benchmark that evaluates large language models on language and writing tasks. LLM Stats tracks 1 model on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.4, with the leader reaching 0.4.

Compare leaders on the best AI for language and best AI for writing leaderboards.

Publication

Paper
RewriteLM: An Instruction-Tuned Large Language Model for Text Rewriting
Authors
Lei Shu, Liangchen Luo, Jayakumar Hoskere, Yun Zhu, and 4 others
Published

Abstract

Large Language Models (LLMs) have demonstrated impressive capabilities in creative tasks such as storytelling and E-mail generation. However, as LLMs are primarily trained on final text results rather than intermediate revisions, it might be challenging for them to perform text rewriting tasks. Most studies in the rewriting tasks focus on a particular transformation type within the boundaries of single sentences. In this work, we develop new strategies for instruction tuning and reinforcement learning to better align LLMs for cross-sentence rewriting tasks using diverse wording and structures expressed through natural languages including 1) generating rewriting instruction data from Wiki edits and public corpus through instruction generation and chain-of-thought prompting; 2) collecting comparison data for reward model training through a new ranking function. To facilitate this research, we introduce OpenRewriteEval, a novel benchmark covers a wide variety of rewriting types expressed through natural language instructions. Our results show significant improvements over a variety of baselines. The public repository is available on GitHub under Google Research (https://github.com/google-research/google-research/tree/master/rewritelm).

MetaLlama 3.2 3B Instruct leads with 40.1%.

Progress Over Time

Interactive timeline showing model performance evolution on Open-rewrite

State-of-the-art frontier
Open
Proprietary

Open-rewrite Leaderboard

1 models
ContextCostLicense
13B
Notice missing or incorrect data?

FAQ

Common questions about Open-rewrite.

What is the Open-rewrite benchmark?

OpenRewriteEval is a benchmark for evaluating open-ended rewriting of long-form texts, covering a wide variety of rewriting types expressed through natural language instructions including formality, expansion, conciseness, paraphrasing, and tone and style transfer.

What is the Open-rewrite leaderboard?

The Open-rewrite leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, Llama 3.2 3B Instruct by Meta leads with a score of 0.401. The average score across all models is 0.401.

What is the highest Open-rewrite score?

The highest Open-rewrite score is 0.401, achieved by Llama 3.2 3B Instruct from Meta.

How many models are evaluated on Open-rewrite?

1 models have been evaluated on the Open-rewrite benchmark, with 0 verified results and 1 self-reported results.

Where can I find the Open-rewrite paper?

The Open-rewrite paper is available at https://arxiv.org/abs/2305.15685. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does Open-rewrite cover?

Open-rewrite is categorized under language and writing. The benchmark evaluates text models.

What is the best open-source model on Open-rewrite?

Llama 3.2 3B Instruct by Meta is the top-ranked open-source model on Open-rewrite, with a score of 0.401 (rank #1).

How recent are the Open-rewrite leaderboard results?

The Open-rewrite leaderboard was last updated in June 2026 and currently includes 1 evaluated models.

More evaluations to explore

Related benchmarks in the same category

View all language
MMLU-Pro

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.

language
127 models
MMLU

Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

language
100 models
MMMLU

Multilingual Massive Multitask Language Understanding dataset released by OpenAI, featuring professionally translated MMLU test questions across 14 languages including Arabic, Bengali, German, Spanish, French, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese, Swahili, Yoruba, and Chinese. Contains approximately 15,908 multiple-choice questions per language covering 57 subjects.

language
48 models
MMLU-Redux

An improved version of the MMLU benchmark featuring manually re-annotated questions to identify and correct errors in the original dataset. Provides more reliable evaluation metrics for language models by addressing dataset quality issues found in the original MMLU.

language
47 models
MMLU-ProX

Extended version of MMLU-Pro providing additional challenging multiple-choice questions for evaluating language models across diverse academic and professional domains. Built on the foundation of the Massive Multitask Language Understanding benchmark framework.

language
30 models
Arena Hard

Arena-Hard-Auto is an automatic evaluation benchmark for instruction-tuned LLMs consisting of 500 challenging real-world prompts curated by BenchBuilder. It includes open-ended software engineering problems, mathematical questions, and creative writing tasks. The benchmark uses LLM-as-a-Judge methodology with GPT-4.1 and Gemini-2.5 as automatic judges to approximate human preference. Arena-Hard achieves 98.6% correlation with human preference rankings and provides 3x higher separation of model performances compared to MT-Bench, making it highly effective for distinguishing between models of similar quality.

writing
26 models