Saved in:
Bibliographic Details
Main Authors: Lim, Junghwan, Lee, Sungmin, Kim, Dongseok, Kim, Taehyun, Park, Eunhwan, Lee, Jeesoo, Lee, Jeongdoo, Lee, Junhyeok, Cheung, Wai Ting, Choi, Dahye, Ha, Minsu, Her, Jaeheui, Huh, Jaeyeon, Jung, Hanbin, Kang, Changjin, Kim, Beomgyu, Kim, Minjae, Kim, Taewhan, Kim, Youngrok, Kweon, Hyukjin, Lee, Haesol, Lee, Kungyu, Oh, Dongpin, Park, Yeongjae, Ryu, Bokki, Weon, Dongjoo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.11463
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909957785387008
author Lim, Junghwan
Lee, Sungmin
Kim, Dongseok
Kim, Taehyun
Park, Eunhwan
Lee, Jeesoo
Lee, Jeongdoo
Lee, Junhyeok
Cheung, Wai Ting
Choi, Dahye
Ha, Minsu
Her, Jaeheui
Huh, Jaeyeon
Jung, Hanbin
Kang, Changjin
Kim, Beomgyu
Kim, Minjae
Kim, Taewhan
Kim, Youngrok
Kweon, Hyukjin
Lee, Haesol
Lee, Kungyu
Oh, Dongpin
Park, Yeongjae
Ryu, Bokki
Weon, Dongjoo
author_facet Lim, Junghwan
Lee, Sungmin
Kim, Dongseok
Kim, Taehyun
Park, Eunhwan
Lee, Jeesoo
Lee, Jeongdoo
Lee, Junhyeok
Cheung, Wai Ting
Choi, Dahye
Ha, Minsu
Her, Jaeheui
Huh, Jaeyeon
Jung, Hanbin
Kang, Changjin
Kim, Beomgyu
Kim, Minjae
Kim, Taewhan
Kim, Youngrok
Kweon, Hyukjin
Lee, Haesol
Lee, Kungyu
Oh, Dongpin
Park, Yeongjae
Ryu, Bokki
Weon, Dongjoo
contents We introduce Motif-2-12.7B-Reasoning, a 12.7B parameter language model designed to bridge the gap between open-weight systems and proprietary frontier models in complex reasoning and long-context understanding. Addressing the common challenges of model collapse and training instability in reasoning adaptation, we propose a comprehensive, reproducible training recipe spanning system, data, and algorithmic optimizations. Our approach combines memory-efficient infrastructure for 64K-token contexts using hybrid parallelism and kernel-level optimizations with a two-stage Supervised Fine-Tuning (SFT) curriculum that mitigates distribution mismatch through verified, aligned synthetic data. Furthermore, we detail a robust Reinforcement Learning Fine-Tuning (RLFT) pipeline that stabilizes training via difficulty-aware data filtering and mixed-policy trajectory reuse. Empirical results demonstrate that Motif-2-12.7B-Reasoning achieves performance comparable to models with significantly larger parameter counts across mathematics, coding, and agentic benchmarks, offering the community a competitive open model and a practical blueprint for scaling reasoning capabilities under realistic compute constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11463
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipes
Lim, Junghwan
Lee, Sungmin
Kim, Dongseok
Kim, Taehyun
Park, Eunhwan
Lee, Jeesoo
Lee, Jeongdoo
Lee, Junhyeok
Cheung, Wai Ting
Choi, Dahye
Ha, Minsu
Her, Jaeheui
Huh, Jaeyeon
Jung, Hanbin
Kang, Changjin
Kim, Beomgyu
Kim, Minjae
Kim, Taewhan
Kim, Youngrok
Kweon, Hyukjin
Lee, Haesol
Lee, Kungyu
Oh, Dongpin
Park, Yeongjae
Ryu, Bokki
Weon, Dongjoo
Artificial Intelligence
We introduce Motif-2-12.7B-Reasoning, a 12.7B parameter language model designed to bridge the gap between open-weight systems and proprietary frontier models in complex reasoning and long-context understanding. Addressing the common challenges of model collapse and training instability in reasoning adaptation, we propose a comprehensive, reproducible training recipe spanning system, data, and algorithmic optimizations. Our approach combines memory-efficient infrastructure for 64K-token contexts using hybrid parallelism and kernel-level optimizations with a two-stage Supervised Fine-Tuning (SFT) curriculum that mitigates distribution mismatch through verified, aligned synthetic data. Furthermore, we detail a robust Reinforcement Learning Fine-Tuning (RLFT) pipeline that stabilizes training via difficulty-aware data filtering and mixed-policy trajectory reuse. Empirical results demonstrate that Motif-2-12.7B-Reasoning achieves performance comparable to models with significantly larger parameter counts across mathematics, coding, and agentic benchmarks, offering the community a competitive open model and a practical blueprint for scaling reasoning capabilities under realistic compute constraints.
title Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipes
topic Artificial Intelligence
url https://arxiv.org/abs/2512.11463