NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model

• URL: http://arxiv.org/abs/2508.14444v4
• Date: Tue, 02 Sep 2025 16:12:36 GMT
• Title: NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
• Abstract summary: Nemotron-Nano-9B-v2 is a hybrid Mamba-Transformer language model designed to increase throughput for reasoning workloads.<n>Nemotron-Nano-9B-v2 builds on the Nemotron-H architecture, in which the majority of the self-attention layers in the common Transformer architecture are replaced with Mamba-2 layers.
• Score: 205.51647932840226
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 compared to similarly-sized models. Nemotron-Nano-9B-v2 builds on the Nemotron-H architecture, in which the majority of the self-attention layers in the common Transformer architecture are replaced with Mamba-2 layers, to achieve improved inference speed when generating the long thinking traces needed for reasoning. We create Nemotron-Nano-9B-v2 by first pre-training a 12-billion-parameter model (Nemotron-Nano-12B-v2-Base) on 20 trillion tokens using an FP8 training recipe. After aligning Nemotron-Nano-12B-v2-Base, we employ the Minitron strategy to compress and distill the model with the goal of enabling inference on up to 128k tokens on a single NVIDIA A10G GPU (22GiB of memory, bfloat16 precision). Compared to existing similarly-sized models (e.g., Qwen3-8B), we show that Nemotron-Nano-9B-v2 achieves on-par or better accuracy on reasoning benchmarks while achieving up to 6x higher inference throughput in reasoning settings like 8k input and 16k output tokens. We are releasing Nemotron-Nano-9B-v2, Nemotron-Nano12B-v2-Base, and Nemotron-Nano-9B-v2-Base checkpoints along with the majority of our pre- and post-training datasets on Hugging Face.

Related papers

• Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence [215.36956014523926]
  Nemotron 3 Nano Omni is the latest model in the Nemotron multimodal series.<n>It is the first to support audio inputs alongside text, images, and video.
  arXiv  Detail & Related papers  (2026-04-27T19:49:21Z)
• Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning [262.0414794728708]
  We describe the pre-training, post-training, and quantization of Nemotron 3 Super.<n>Nemotron 3 Super is a 120 billion (active 12 billion) parameter hybrid Mamba-Attention Mixture-of-Experts model.
  arXiv  Detail & Related papers  (2026-04-14T07:02:32Z)
• NVIDIA Nemotron 3: Efficient and Open Intelligence [227.47413816066845]
  Nemotron 3 family of models deliver strong agentic, reasoning, and conversational capabilities.<n>Nemotron 3 models are post-trained using multi-environment reinforcement learning enabling reasoning, multi-step tool use, and support granular reasoning budget control.<n>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.
  arXiv  Detail & Related papers  (2025-12-24T00:24:05Z)
• Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning [223.88531320605486]
  Nemotron 3 Nano 30B-A3B is a Mixture-of-Experts hybrid Mamba-Transformer language model.<n>Nemotron 3 Nano was pretrained on 25 trillion text tokens, including more than 3 trillion new unique tokens over Nemotron 2.
  arXiv  Detail & Related papers  (2025-12-23T23:54:32Z)
• NVIDIA Nemotron Nano V2 VL [140.7011776408786]
  Nemotron Nano V2 VL builds on Nemotron Nano V2, a hybrid Mamba-Transformer LLM.<n>We are releasing model checkpoints in BF16, FP8, and FP4 formats.
  arXiv  Detail & Related papers  (2025-11-06T00:10:19Z)
• Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search [42.46046429414803]
  Jet-Nemotron is a new family of hybrid-architecture language models.<n>It matches or exceeds the accuracy of leading full-attention models.
  arXiv  Detail & Related papers  (2025-08-21T17:59:08Z)
• Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models [164.47008715747822]
  Nemotron-H is a family of 8B and 56B/47B hybrid Mamba-Transformer models.<n>We replace the majority of self-attention layers in the common Transformer model architecture with Mamba layers.<n>Nemotron-H models offer either better or on-par accuracy compared to other similarly-sized open-sourced Transformer models.
  arXiv  Detail & Related papers  (2025-04-04T17:41:58Z)
• The Mamba in the Llama: Distilling and Accelerating Hybrid Models [76.64055251296548]
  We show how to distill large Transformers into linear RNNs by reusing the linear projection weights from attention layers with academic GPU resources.<n>The resulting hybrid model achieves performance comparable to the original Transformer in chat benchmarks.<n>We also introduce a hardware-aware speculative decoding algorithm that accelerates the inference speed of Mamba and hybrid models.
  arXiv  Detail & Related papers  (2024-08-27T17:56:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.