832 citations · 1.9k across the 80 of their papers we have counts for
57 papers · 1 filter
Unified Audio Intelligence Without Regressing on Text Intelligence
Zhifeng Kong, Sang-gil Lee, Jaehyeon Kim +17
Audio intelligence involves understanding, reasoning about, and generating both audio and speech. In this work, we introduce Nemotron-Labs-Audex-30B-A3B (Audex), a unified audio-te…
Nemotron-Labs-TwoTower: Diffusion Language Modeling with Pretrained Autoregressive Context
Fitsum Reda, John Kamalu, Roger Waleffe +3
Diffusion language models offer a promising alternative to autoregressive models due to their potential for parallel and iterative generation. However, existing approaches use a si…
Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +572
We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 t…
Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
Zhuolin Yang, Zihan Liu, Yang Chen +14
We introduce Nemotron-Cascade 2, an open 30B MoE model with 3B activated parameters that delivers best-in-class reasoning and strong agentic capabilities. Despite its compact size,…
MMOU: A Massive Multi-Task Omni Understanding and Reasoning Benchmark for Long and Complex Real-World Videos
Arushi Goel, Sreyan Ghosh, Vatsal Agarwal +16
Multimodal Large Language Models (MLLMs) have shown strong performance in visual and audio understanding when evaluated in isolation. However, their ability to jointly reason over…
On Data Engineering for Scaling LLM Terminal Capabilities
Renjie Pi, Grace Lam, Mohammad Shoeybi +3
Despite rapid recent progress in the terminal capabilities of large language models, the training data strategies behind state-of-the-art terminal agents remain largely undisclosed…
NVIDIA Nemotron 3: Efficient and Open Intelligence
NVIDIA, :, Aaron Blakeman +356
We introduce the Nemotron 3 family of models - Nano, Super, and Ultra. These models deliver strong agentic, reasoning, and conversational capabilities. The Nemotron 3 family uses a…
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +311
We present Nemotron 3 Nano 30B-A3B, a Mixture-of-Experts hybrid Mamba-Transformer language model. Nemotron 3 Nano was pretrained on 25 trillion text tokens, including more than 3 t…
Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
Boxin Wang, Chankyu Lee, Nayeon Lee +9
Building general-purpose reasoning models with reinforcement learning (RL) entails substantial cross-domain heterogeneity, including large variation in inference-time response leng…
Nemotron Elastic: Towards Efficient Many-in-One Reasoning LLMs
Ali Taghibakhshi, Sharath Turuvekere Sreenivas, Saurav Muralidharan +13
Training a family of large language models targeting multiple scales and deployment objectives is prohibitively expensive, requiring separate training runs for each different size.…
NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
NVIDIA, :, Aarti Basant +214
We introduce Nemotron-Nano-9B-v2, a hybrid Mamba-Transformer language model designed to increase throughput for reasoning workloads while achieving state-of-the-art accuracy compar…
Pretraining Large Language Models with NVFP4
NVIDIA, Felix Abecassis, Anjulie Agrusa +87
Large Language Models (LLMs) today are powerful problem solvers across many domains, and they continue to get stronger as they scale in model size, training set size, and training…
Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset
Rabeeh Karimi Mahabadi, Sanjeev Satheesh, Shrimai Prabhumoye +3
Pretraining large language models (LLMs) on high-quality, structured data such as mathematics and code substantially enhances reasoning capabilities. However, existing math-focused…
AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy
Zihan Liu, Zhuolin Yang, Yang Chen +4
In this work, we investigate the synergy between supervised fine-tuning (SFT) and reinforcement learning (RL) in developing strong reasoning models. We begin by curating the SFT tr…
From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models
Chejian Xu, Wei Ping, Peng Xu +5
Long-context capabilities are essential for a wide range of applications, including document and video understanding, in-context learning, and inference-time scaling, all of which…
Minitron-SSM: Efficient Hybrid Language Model Compression through Group-Aware SSM Pruning
Ali Taghibakhshi, Sharath Turuvekere Sreenivas, Saurav Muralidharan +15
Hybrid LLM architectures that combine Attention and State Space Models (SSMs) achieve state-of-the-art accuracy and runtime performance. Recent work has demonstrated that applying…
Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models
NVIDIA, :, Aaron Blakeman +198
As inference-time scaling becomes critical for enhanced reasoning capabilities, it is increasingly becoming important to build models that are efficient to infer. We introduce Nemo…
AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modeling
Zihan Liu, Yang Chen, Mohammad Shoeybi +2
In this paper, we introduce AceMath, a suite of frontier math models that excel in solving complex math problems, along with highly effective reward models capable of evaluating ge…
Maximize Your Data's Potential: Enhancing LLM Accuracy with Two-Phase Pretraining
Steven Feng, Shrimai Prabhumoye, Kezhi Kong +4
Pretraining large language models effectively requires strategic data selection, blending and ordering. However, key details about data mixtures especially their scalability to lon…
Nemotron-CC: Transforming Common Crawl into a Refined Long-Horizon Pretraining Dataset
Dan Su, Kezhi Kong, Ying Lin +6
Recent English Common Crawl datasets like FineWeb-Edu and DCLM achieved significant benchmark gains via aggressive model-based filtering, but at the cost of removing 90% of data. T…
MM-Embed: Universal Multimodal Retrieval with Multimodal LLMs
Sheng-Chieh Lin, Chankyu Lee, Mohammad Shoeybi +3
State-of-the-art retrieval models typically address a straightforward search scenario, in which retrieval tasks are fixed (e.g., finding a passage to answer a specific question) an…
NVLM: Open Frontier-Class Multimodal LLMs
Wenliang Dai, Nayeon Lee, Boxin Wang +7
We introduce NVLM 1.0, a family of frontier-class multimodal large language models (LLMs) that achieve state-of-the-art results on vision-language tasks, rivaling the leading propr…
Upcycling Large Language Models into Mixture of Experts
Ethan He, Abhinav Khattar, Ryan Prenger +7
Upcycling pre-trained dense language models into sparse mixture-of-experts (MoE) models is an efficient approach to increase the model capacity of already trained models. However,…
LLM Pruning and Distillation in Practice: The Minitron Approach
Sharath Turuvekere Sreenivas, Saurav Muralidharan, Raviraj Joshi +19
We present a comprehensive report on compressing the Llama 3.1 8B and Mistral NeMo 12B models to 4B and 8B parameters, respectively, using pruning and distillation. We explore two…