* The post-training data has a cutoff date of November 28, 2025\. * The pre-training data has a cutoff date of June 25, 2025\. Nemotron-Nano-3-30B-A3B-FP8 is a quantized version of Nemotron-Nano-3-30B-A3B and is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be configured through a flag in the chat template. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be configured to do so, albeit with a slight decrease in accuracy for harder prompts that require reasoning. Conversely, allowing the model to generate reasoning traces first generally results in higher-quality final solutions to queries and tasks. The model employs a hybrid Mixture-of-Experts (MoE) architecture, consisting of 23 Mamba-2 and MoE layers, along with 6 Attention layers. Each MoE layer includes 128 experts plus 1 shared expert, with 6 experts activated per token. The model has 3.5B active parameters and 30B parameters in total.
* The post-training data has a cutoff date of November 28, 2025\. * The pre-training data has a cutoff date of June 25, 2025\. Nemotron-Nano-3-30B-A3B-NVFP4 is a quantized version of Nemotron-Nano-3-30B-A3B and is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be configured through a flag in the chat template. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be configured to do so, albeit with a slight decrease in accuracy for harder prompts that require reasoning. Conversely, allowing the model to generate reasoning traces first generally results in higher-quality final solutions to queries and tasks. The model employs a hybrid Mixture-of-Experts (MoE) architecture, consisting of 23 Mamba-2 and MoE layers, along with 6 Attention layers. Each MoE layer includes 128 experts plus 1 shared expert, with 6 experts activated per token. The model has 3.5B active parameters and 30B parameters in total.
NVIDIA-Nemotron-3-Nano-4B-BF16 is a small language model (SLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be controlled via a system prompt. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be configured to do so, albeit with a slight decrease in accuracy for harder prompts that require reasoning. Conversely, allowing the model to generate reasoning traces first generally results in higher-quality final solutions to queries and tasks. The model has been compressed from NVIDIA-Nemotron-Nano-9B-v2 using the Nemotron Elastic framework. The details of the parent model NVIDIA-Nemotron-Nano-9B-v2 can be found in (Nemotron-H tech report). The model uses a hybrid architecture consisting primarily of Mamba-2 and MLP layers combined with just four Attention layers. Governing Terms: Use of this model is governed by the NVIDIA Nemotron Open Model License.
NVIDIA-Nemotron-3-Nano-4B-FP8 is quantized version of NVIDIA-Nemotron-3-Nano-4B-BF16 and is a small language model (SLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be controlled via a system prompt. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be configured to do so, albeit with a slight decrease in accuracy for harder prompts that require reasoning. Conversely, allowing the model to generate reasoning traces first generally results in higher-quality final solutions to queries and tasks. The model has been compressed from NVIDIA-Nemotron-Nano-9B-v2 using the Nemotron Elastic framework. The details of the parent model NVIDIA-Nemotron-Nano-9B-v2 can be found in (Nemotron-H tech report). The model uses a hybrid architecture consisting primarily of Mamba-2 and MLP layers combined with just four Attention layers. Governing Terms: Use of this model is governed by the NVIDIA Nemotron Open Model License.
NVIDIA-Nemotron-3-Nano-4B-GGUF is quantized version (Q4_K_M) of NVIDIA-Nemotron-3-Nano-4B-BF16 and is a small language model (SLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be controlled via a system prompt. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be configured to do so, albeit with a slight decrease in accuracy for harder prompts that require reasoning. Conversely, allowing the model to generate reasoning traces first generally results in higher-quality final solutions to queries and tasks. The model has been compressed from NVIDIA-Nemotron-Nano-9B-v2 using the Nemotron Elastic framework. The details of the parent model NVIDIA-Nemotron-Nano-9B-v2 can be found in (Nemotron-H tech report). The model uses a hybrid architecture consisting primarily of Mamba-2 and MLP layers combined with just four Attention layers. Governing Terms: Use of this model is governed by the NVIDIA Nemotron Open Model License.
**Nemotron-3-Super-120B-A12B-Base** is a base large language model (LLM) trained from scratch by NVIDIA, with next token prediction loss. It provides a good starting platform for further training, including instruction follow, and coding. The model employs a hybrid **Latent Mixture-of-Experts (LatentMoE)** architecture, utilizing interleaved Mamba-2 and MoE layers, along with select Attention layers. Distinct from the Nano model, the Super model incorporates **Multi-Token Prediction (MTP)** layers for faster text generation and improved quality, and it is trained using **NVFP4** quantization to maximize compute efficiency. The model has **12B active parameters** and **120B parameters in total**. The supported languages include: English, Spanish, French, German, Japanese, Italian, Chinese, Arabic, Hebrew, Hindi, Korean, Czech, Danish, Dutch, Finnish, Polish, Portuguese, Thai, Swedish, and Russian.
> Use `temperature=1.0` and `top_p=0.95` across **all tasks and serving backends** — reasoning, tool calling, and general chat alike. For more details on how to deploy and use the model - see the Quick Start Guide below! > *For running Nemotron 3 Super on a single B200 or DGX Spark - please see: NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4*
> Use `temperature=1.0` and `top_p=0.95` across **all tasks and serving backends** — reasoning, tool calling, and general chat alike. For more details on how to deploy and use the model - see the Quick Start Guide below! * The post-training data has a cutoff date of February 2026. * The pre-training data has a cutoff date of June 2025.
For more details on how to deploy and use the model - see the Quick Start Guide below! > *For running Nemotron 3 Ultra on a smaller footprint, please see: NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4* * The post-training data has a cutoff date of May 2026. * The pre-training data has a cutoff date of September 2025.
For more details on how to deploy and use the model - see the Quick Start Guide below! * The post-training data has a cutoff date of May 2026. * The pre-training data has a cutoff date of September 2025. NVIDIA Nemotron™ is a family of open models with open weights, training data, and recipes, delivering leading efficiency and accuracy for building specialized AI agents.
**NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16** is a looped (latent-recurrent) large language model (LLM) trained by NVIDIA. The model employs a normalized-GPT architecture with interleaved hybrid sliding window attention (SWA) and full attention layers. By repeatedly applying a shared block of layers, Dory1.0-4B can spend more compute per token allowing it to scale intelligence. This model is ready for commercial or non-commercial use. Dory (Looped LM) Architecture: The layer stack is divided into three groups: Input (I), Recurrent (R) and Output (O). The shared Recurrent block (R) is applied multiple times between the Input and Output blocks, and the input is reinjected into every loop. In Dory, the Input and Output blocks have 8 layers each, while the Recurrent block has 20 layers. Accuracy numbers measured by NVIDIA under a consistent harness (NeMo Gym / Nemo Evaluator SDK). Terminal-Bench 2.1 uses the Terminus-2 harness for all models. † Reported scores.
