Apriel-Nemotron-15B-Thinker
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909737118859264 |
|---|---|
| author | Radhakrishna, Shruthan Parikh, Soham Sarda, Gopal Turkkan, Anil Vohra, Quaizar Li, Raymond Jhamb, Dhruv Ogueji, Kelechi Shukla, Aanjaneya Bamgbose, Oluwanifemi Liang, Toby Kumar, Luke Ostapenko, Oleksiy Malay, Shiva Krishna Reddy Tiwari, Aman Bogavelli, Tara Yadav, Vikas Mehta, Jash Mittal, Saloni Kalkunte, Akshay Pattnaik, Pulkit Slimi, Khalil Sreeram, Anirudh Nair, Jishnu Oladipo, Akintunde Maiya, Shashank Mahajan, Khyati Maheshwary, Rishabh Hashemi, Masoud Mudumba, Sai Rajeswar Madhusudhan, Sathwik Tejaswi Scholak, Torsten Paquet, Sebastien Davasam, Sagar Sunkara, Srinivas |
| author_facet | Radhakrishna, Shruthan Parikh, Soham Sarda, Gopal Turkkan, Anil Vohra, Quaizar Li, Raymond Jhamb, Dhruv Ogueji, Kelechi Shukla, Aanjaneya Bamgbose, Oluwanifemi Liang, Toby Kumar, Luke Ostapenko, Oleksiy Malay, Shiva Krishna Reddy Tiwari, Aman Bogavelli, Tara Yadav, Vikas Mehta, Jash Mittal, Saloni Kalkunte, Akshay Pattnaik, Pulkit Slimi, Khalil Sreeram, Anirudh Nair, Jishnu Oladipo, Akintunde Maiya, Shashank Mahajan, Khyati Maheshwary, Rishabh Hashemi, Masoud Mudumba, Sai Rajeswar Madhusudhan, Sathwik Tejaswi Scholak, Torsten Paquet, Sebastien Davasam, Sagar Sunkara, Srinivas |
| contents | While large language models (LLMs) have achieved remarkable reasoning capabilities across domains like code, math and other enterprise tasks, their significant memory and computational costs often preclude their use in practical enterprise settings. To this end, we introduce Apriel-Nemotron-15B-Thinker, a 15-billion parameter model in the ServiceNow Apriel SLM series that achieves performance against medium sized state-of-the-art models such as o1-mini, QWQ32B, and EXAONE-Deep-32B while maintaining only half the memory footprint of those alternatives. Apriel-Nemotron-15B-Thinker model is trained in a four stage training pipeline including 1) Base Model upscaling, 2) Continual Pre-training 3) Supervised Fine-tuning (SFT) and 4) Reinforcement Learning using GRPO. Comprehensive evaluations across a diverse suite of benchmarks consistently demonstrate that our Apriel-Nemotron-15B-Thinker model matches or exceeds the performance of its 32-billion parameter counterparts, despite being less than half their size. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_10948 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Apriel-Nemotron-15B-Thinker Radhakrishna, Shruthan Parikh, Soham Sarda, Gopal Turkkan, Anil Vohra, Quaizar Li, Raymond Jhamb, Dhruv Ogueji, Kelechi Shukla, Aanjaneya Bamgbose, Oluwanifemi Liang, Toby Kumar, Luke Ostapenko, Oleksiy Malay, Shiva Krishna Reddy Tiwari, Aman Bogavelli, Tara Yadav, Vikas Mehta, Jash Mittal, Saloni Kalkunte, Akshay Pattnaik, Pulkit Slimi, Khalil Sreeram, Anirudh Nair, Jishnu Oladipo, Akintunde Maiya, Shashank Mahajan, Khyati Maheshwary, Rishabh Hashemi, Masoud Mudumba, Sai Rajeswar Madhusudhan, Sathwik Tejaswi Scholak, Torsten Paquet, Sebastien Davasam, Sagar Sunkara, Srinivas Machine Learning Artificial Intelligence While large language models (LLMs) have achieved remarkable reasoning capabilities across domains like code, math and other enterprise tasks, their significant memory and computational costs often preclude their use in practical enterprise settings. To this end, we introduce Apriel-Nemotron-15B-Thinker, a 15-billion parameter model in the ServiceNow Apriel SLM series that achieves performance against medium sized state-of-the-art models such as o1-mini, QWQ32B, and EXAONE-Deep-32B while maintaining only half the memory footprint of those alternatives. Apriel-Nemotron-15B-Thinker model is trained in a four stage training pipeline including 1) Base Model upscaling, 2) Continual Pre-training 3) Supervised Fine-tuning (SFT) and 4) Reinforcement Learning using GRPO. Comprehensive evaluations across a diverse suite of benchmarks consistently demonstrate that our Apriel-Nemotron-15B-Thinker model matches or exceeds the performance of its 32-billion parameter counterparts, despite being less than half their size. |
| title | Apriel-Nemotron-15B-Thinker |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2508.10948 |