mlsanityData to 13 Oct 2026

NVIDIA Nemotron 3: Efficient and Open Intelligence

arXiv 2512.20856 ↗First posted 24 Dec 2025cs.CL cs.AI cs.LGcited by 57

Not on the map. Why?

Cited by 22 in LLM reasoning, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

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 Mixture-of-Experts hybrid Mamba-Transformer architecture to provide best-in-class throughput and context lengths of up to 1M tokens. Super and Ultra models are trained with NVFP4 and incorporate LatentMoE, a novel approach that improves model quality. The two larger models also include MTP layers for faster text generation. All Nemotron 3 models are post-trained using multi-environment reinforcement learning enabling reasoning, multi-step tool use, and support granular reasoning budget control. Nano, the smallest model, outperforms comparable models in accuracy while remaining extremely cost-efficient for inference. Super is optimized for collaborative agents and high-volume workloads such as IT ticket automation. Ultra, the largest model, provides state-of-the-art accuracy and reasoning performance. Nano is released together with its technical report and this white paper, while Super and Ultra will follow in the coming months. We will openly release the model weights, pre- and post-training software, recipes, and all data for which we hold redistribution rights.

Key facts. Cited by 57 papers in the map (data through 13 Oct 2026). Builds on 9 of its 12 references that are in the map. First posted on arXiv 24 Dec 2025.

Builds on

9 of 12 references.

References · 9

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,091
  2. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,625
  3. 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,307
  4. 2024RULER: What's the Real Context Size of Your Long-Context Language Models?cited by 1,302
  5. 2024Better & Faster Large Language Models via Multi-token Predictioncited by 350
  6. 2024An Empirical Study of Mamba-based Language Modelscited by 179
  7. 2025NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Modelcited by 99
  8. 2025Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Modelscited by 93
  9. 2025Pretraining Large Language Models with NVFP4cited by 71

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 57.

57 papers.

Sort:

By area:

  1. 2026Cosmos 3: Omnimodal World Models for Physical AIcited by 153
  2. 2026Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agentscited by 57
  3. 2026Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoningcited by 36
  4. 2026Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoningcited by 34
  5. 2026Lightning OPD: Efficient Post-Training for Large Reasoning Models with Offline On-Policy Distillationcited by 31
  6. 2026Scalable Training of Mixture-of-Experts Models with Megatron Corecited by 19
  7. 2026Cross-Model KV Cache Transfer in LLM Families: A Closed-Form Linear Mapping for Prefill Reusecited by 10
  8. 2026SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decodingcited by 8
  9. 2026M2RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modelingcited by 7
  10. 2026Hardware Co-Design Scaling Laws via Roofline Modelling for On-Device LLMscited by 5
  11. 2026ReasonXL: Shifting LLM Reasoning Language Without Sacrificing Performancecited by 5
  12. 2026Evaluation Awareness in Language Models Has Limited Effect on Behaviourcited by 5
38 more
  1. 2026FLARE: Diffusion for Hybrid Language Modelcited by 5
  2. 2026Rethinking the Role of Efficient Attention in Hybrid Architecturescited by 5
  3. 2026Nudging Beyond the Comfort Zone: Efficient Strategy-Guided Exploration for RLVRcited by 4
  4. 2026Raven: High-Recall Sequence Modeling with Sparse Memory Routingcited by 4
  5. 2026Lightning OPD 2.0: Mitigating Style Bias in Cross-Teacher On-Policy Distillation for Large Reasoning Modelscited by 4
  6. 2026An Empirical Study of Harness Design for Coding Agentscited by 4
  7. 2026Evaluating the Progression of Large Language Model Capabilities for Small-Molecule Drug Designcited by 3
  8. 2026Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal Statescited by 3
  9. 2026Nemotron-Labs-TwoTower: Diffusion Language Modeling with Pretrained Autoregressive Contextcited by 3
  10. 2026Hidden State Poisoning Attacks against Mamba-based Language Modelscited by 2
  11. 2026OpenClaw-Skill: Collective Skill Tree Search for Agentic Large Language Modelscited by 2
  12. 2026Judging What We Cannot Solve: A Consequence-Based Approach for Oracle-Free Evaluation of Research-Level Mathcited by 1
  13. 2026Extending Puzzle for Mixture-of-Experts Reasoning Models with Application to GPT-OSS Accelerationcited by 1
  14. 2026Mind the Gap Between Spatial Reasoning and Acting! Step-by-Step Evaluation of Agents With Spatial-Gymcited by 1
  15. 2026Introspective X Training: Feedback Conditioning Improves Scaling Across all LLM Training Stagescited by 1
  16. 2026Which Models Are Our Models Built On? Auditing Invisible Dependencies in Modern LLMscited by 1
  17. 2026Nemotron-Labs-3-Puzzle-75B-A9B: Compressing Hybrid MoE LLMscited by 1
  18. 2026Synthetic Persona Pretraining: Alignment from Token Zerocited by 1
  19. 2026STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughtscited by 0
  20. 2026Swimba: Switch Mamba Model Scales State Space Modelscited by 0
  21. 2026Jagle: Building a Large-Scale Japanese Multimodal Post-Training Dataset for Vision-Language Modelscited by 0
  22. 2026How Transformers Learn to Plan via Multi-Token Predictioncited by 0
  23. 2026When Reasoning Supervision Hurts: TTCW-Based Long-Form Literary Review Generationcited by 0
  24. 2026DASH: Fast Differentiable Architecture Search for Hybrid Attention in Minutes on a Single GPUcited by 0
  25. 2026The Epi-LLM Framework: probing LLM behavioral priors through epidemiological agent-based modelscited by 0
  26. 2026UnpredictaBench: A Benchmark for Evaluating Distributional Randomness in LLMscited by 0
  27. 2026AFFORDANCE20Q: Evaluating Affordance Reasoning from Physical Propertiescited by 0
  28. 2026ThinkProbe: Beyond Accuracy -- Structural Profiling of Open-Ended LLM Reasoning Traces via Non-Generative Thought Graphscited by 0
  29. 2026Thinking Under Uncertainty: Evidence Use and Information-Seeking in Language Modelscited by 0
  30. 2026EarlyDx: An Admission-Anchored Benchmark for Open-Ended Generation of Evidence-Supported ED-Encounter Diagnosescited by 0
  31. 2026Unified Multi-Dimensional Benchmark for Complex Graph Reasoning in Large Language Modelscited by 0
  32. 2026Reasoning Jury: Multi-Model Consensus for Evaluating Reasoning Tracescited by 0
  33. 2026Training Communication-Efficient Mixture-of-Experts Language Models with Layer Re-Configurationcited by 0
  34. 2026Safety Signals to Verify NetOps Agents with Action-Level Granularitycited by 0
  35. 2026Bellman Policy Optimizationcited by 0
  36. 2026On the Importance of Gating: Memorization vs. In-Context Learning in State Space Modelscited by 0
  37. 2026Strategically Diverse Sampling for Self-Trainingcited by 0
  38. 2026Counting on Thinking: Tracing Evidence Integration in Language Modelscited by 0

50 most cited of 57 All 57 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
  2. 2025Llama-Nemotron: Efficient Reasoning Models
  3. 2026Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence
  4. 2025NVIDIA Nemotron Nano V2 VL
  5. 2026Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
  6. 2025Apriel-Nemotron-15B-Thinker
  7. 2025Nemotron Elastic: Towards Efficient Many-in-One Reasoning LLMs
  8. 2025Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search