mlsanityData to 7 Oct 2026

Llama-Nemotron: Efficient Reasoning Models

arXiv 2505.00949 ↗First posted 2 May 2025cs.CL cs.AI cs.LGcited by 151

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

We introduce the Llama-Nemotron series of models, an open family of heterogeneous reasoning models that deliver exceptional reasoning capabilities, inference efficiency, and an open license for enterprise use. The family comes in three sizes -- Nano (8B), Super (49B), and Ultra (253B) -- and performs competitively with state-of-the-art reasoning models such as DeepSeek-R1 while offering superior inference throughput and memory efficiency. In this report, we discuss the training procedure for these models, which entails using neural architecture search from Llama 3 models for accelerated inference, knowledge distillation, and continued pretraining, followed by a reasoning-focused post-training stage consisting of two main parts: supervised fine-tuning and large scale reinforcement learning. Llama-Nemotron models are the first open-source models to support a dynamic reasoning toggle, allowing users to switch between standard chat and reasoning modes during inference. To further support open research and facilitate model development, we provide the following resources: 1. We release the Llama-Nemotron reasoning models -- LN-Nano, LN-Super, and LN-Ultra -- under the commercially permissive NVIDIA Open Model License Agreement. 2. We release the complete post-training dataset: Llama-Nemotron-Post-Training-Dataset. 3. We also release our training codebases: NeMo, NeMo-Aligner, and Megatron-LM.

Key facts. Cited by 151 papers in the map (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 25 of its 41 references that are in the map. First posted on arXiv 2 May 2025.

Builds on

25 of 41 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025s1: Simple test-time scaling31 Jan

Other references · 24

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  3. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  5. 2023Let's Verify Step by Stepcited by 3,598
  6. 2021Program Synthesis with Large Language Modelscited by 3,509
  7. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  8. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  9. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
  10. 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,829
  11. 2022Competition-Level Code Generation with AlphaCodecited by 1,427
  12. 2023GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpointscited by 1,311
12 more
  1. 2023Instruction-Following Evaluation for Large Language Modelscited by 1,143
  2. 2023SGLang: Efficient Execution of Structured Language Model Programscited by 1,107
  3. 2021Measuring Coding Challenge Competence With APPScited by 879
  4. 2024JudgeBench: A Benchmark for Evaluating LLM-based Judgescited by 294
  5. 2024HelpSteer2: Open-source dataset for training top-performing reward modelscited by 232
  6. 2024OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Datacited by 194
  7. 2023Rethinking Benchmark and Contamination for Language Models with Rephrased Samplescited by 192
  8. 2025AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning datasetcited by 165
  9. 2023TACO: Topics in Algorithmic COde generation datasetcited by 150
  10. 2025OpenCodeReasoning: Advancing Data Distillation for Competitive Codingcited by 106
  11. 2024NeMo-Aligner: Scalable Toolkit for Efficient Model Alignmentcited by 40
  12. 2025Reward-aware Preference Optimization: A Unified Mathematical Framework for Model Alignmentcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 79; 2026: 72.

151 papers.

Sort:

By area:

  1. 2025OpenThoughts: Data Recipes for Reasoning Modelscited by 287
  2. 2025R-Zero: Self-Evolving Reasoning LLM from Zero Datacited by 210
  3. 2025Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergycited by 204
  4. 2025Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontierscited by 193
  5. 2025DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generationcited by 184
  6. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 175
  7. 2025Scaling Latent Reasoning via Looped Language Modelscited by 145
  8. 2025OctoThinker: Mid-training Incentivizes Reinforcement Learning Scalingcited by 110
  9. 2025NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Modelcited by 94
  10. 2025Curriculum Reinforcement Learning from Easy to Hard Tasks Improves LLM Reasoningcited by 89
  11. 2025AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learningcited by 85
  12. 2025AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergycited by 75
38 more
  1. 2025Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognitioncited by 61
  2. 2025MegaScience: Pushing the Frontiers of Post-Training Datasets for Science Reasoningcited by 56
  3. 2025Training a Scientific Reasoning Model for Chemistrycited by 54
  4. 2025rStar2-Agent: Agentic Reasoning Technical Reportcited by 54
  5. 2025Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodalitycited by 49
  6. 2025Beyond Chemical QA: Evaluating LLM's Chemical Reasoning with Modular Chemical Operationscited by 46
  7. 2025Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Modelscited by 45
  8. 2025ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestrationcited by 44
  9. 2025ThinkSwitcher: When to Think Hard, When to Think Fastcited by 37
  10. 2025A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontierscited by 34
  11. 2026d3LLM: Ultra-Fast Diffusion LLM using Pseudo-Trajectory Distillationcited by 34
  12. 2025Nudging the Boundaries of LLM Reasoningcited by 25
  13. 2025OpenCodeReasoning-II: A Simple Test Time Scaling Approach via Self-Critiquecited by 24
  14. 2025Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Insteadcited by 24
  15. 2025Llama-3.1-FoundationAI-SecurityLLM-8B-Instruct Technical Reportcited by 21
  16. 2026Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generationcited by 17
  17. 2025DisCO: Reinforcing Large Reasoning Models with Discriminative Constrained Optimizationcited by 16
  18. 2025Intrinsic Fingerprint of LLMs: Continue Training is NOT All You Need to Steal A Model!cited by 16
  19. 2025Safety Through Reasoning: An Empirical Study of Reasoning Guardrail Modelscited by 14
  20. 2025Adaptive Deep Reasoning: Triggering Deep Thinking When Neededcited by 14
  21. 2025Knowledge or Reasoning? A Close Look at How LLMs Think Across Domainscited by 14
  22. 2025Multipole Attention for Efficient Long Context Reasoningcited by 14
  23. 2025PhysUniBench: A Multi-Modal Physics Reasoning Benchmark at Undergraduate Levelcited by 14
  24. 2025HoneyBee: Data Recipes for Vision-Language Reasonerscited by 13
  25. 2026Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoningcited by 13
  26. 2025REST: Stress Testing Large Reasoning Models by Asking Multiple Problems at Oncecited by 12
  27. 2025Energy Use of AI Inference, Efficiency Pathways, and Test-Time Scalingcited by 12
  28. 2025Through the Valley: Path to Effective Long CoT Training for Small Language Modelscited by 10
  29. 2025Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thoughtcited by 10
  30. 2025UALM: Unified Audio Language Model for Understanding, Generation and Reasoningcited by 10
  31. 2025Unforgotten Safety: Preserving Safety Alignment of Large Language Models with Continual Learningcited by 10
  32. 2026Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignmentcited by 10
  33. 2026RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Modelscited by 9
  34. 2025ToTRL: Unlock LLM Tree-of-Thoughts Reasoning Potential through Puzzles Solvingcited by 8
  35. 2025On The Fragility of Benchmark Contamination Detection in Reasoning Modelscited by 8
  36. 2025Webscale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining Levelscited by 8
  37. 2025Shape of Thought: When Distribution Matters More than Correctness in Reasoning Taskscited by 8
  38. 2025How and Why LLMs Generalize: A Fine-Grained Analysis of LLM Reasoning from Cognitive Behaviors to Low-Level Patternscited by 8

50 most cited of 151 All 151 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
  2. 2026Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
  3. 2025Nemotron Elastic: Towards Efficient Many-in-One Reasoning LLMs
  4. 2025Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models
  5. 2024Assessing the Emergent Symbolic Reasoning Abilities of Llama Large Language Models
  6. 2026Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
  7. 2025NVIDIA Nemotron 3: Efficient and Open Intelligence
  8. 2025Nemotron-CrossThink: Scaling Self-Learning beyond Math Reasoning