# Open R1: Update #3

DevFeed: [Open R1: Update #3](<https://devfeed.tech/articles/open-r1-update-3-7423.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/open-r1/update-3>)

Author: Guilherme Penedo; Lewis Tunstall; Anton Lozhkov; Hynek Kydlicek; Edward Beeching; Loubna Ben Allal; Quentin Gallouédec; Leandro von Werra; Agustín Piqueres Lajarín; Nathan Habib

Published: 2025-03-11T20:40:47Z

Content type: article

Language: en

Sources: [Hugging Face - Blog](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [Programming](<https://devfeed.tech/topics/programming.md>), [Code](<https://devfeed.tech/topics/code.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [dataset](<https://devfeed.tech/topics/dataset.md>), [C++](<https://devfeed.tech/topics/c-plus-plus.md>), [Python](<https://devfeed.tech/topics/python.md>), [Chain-of-thought](<https://devfeed.tech/topics/chain-of-thought.md>), [deepseek](<https://devfeed.tech/topics/deepseek.md>)

Tags: [2025](<https://devfeed.tech/tags/2025.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [c-plus-plus](<https://devfeed.tech/tags/c-plus-plus.md>), [chain-of-thought](<https://devfeed.tech/tags/chain-of-thought.md>), [code](<https://devfeed.tech/tags/code.md>), [contests](<https://devfeed.tech/tags/contests.md>), [data](<https://devfeed.tech/tags/data.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [performance](<https://devfeed.tech/tags/performance.md>), [programming](<https://devfeed.tech/tags/programming.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [solutions](<https://devfeed.tech/tags/solutions.md>), [training](<https://devfeed.tech/tags/training.md>), [update](<https://devfeed.tech/tags/update.md>)

## AI overview

Open R1 Update #3 presents CodeForces-CoTs, a dataset of nearly 100,000 DeepSeek-R1 chain-of-thought samples for generating competitive-programming solutions in C++ and Python. It also introduces the IOI benchmark and OlympicCoder models fine-tuned on this data, reporting strong performance on challenging IOI problems.

## Source excerpt

Over the last few weeks, we have focused our efforts on reproducing the competitive programming (code reasoning) aspects of the DeepSeek-R1 recipe. In this post, we are excited to share: - The construction of CodeForces-CoTs: a dataset of nearly 100k high-quality samples distilled from R1 to produce solutions in C++ and Python. - The IOI benchmark: a new benchmark of challenging problems from the 2024 International Olympiad in Informatics (IOI).