Nemotron 3 Nano \- A new Standard for Efficient, Open, and Intelligent Agentic Models
•
64
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching
SpaceTools: Tool-Augmented Spatial Reasoning via Double Interactive RL
Articles
Open, Production-ready Enterprise Models
-
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16
Text Generation • 32B • Updated • 315 • 49 -
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
Text Generation • 32B • Updated • 10.5k • 250 -
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8
Text Generation • 32B • Updated • 12.7k • 94 -
nvidia/Qwen3-Nemotron-235B-A22B-GenRM
Text Generation • 235B • Updated • 48 • 6
Collection of datasets used in the post-training phase of Nemotron Nano v3.
Open, Production-ready Enterprise Models. Nvidia Open Model license.
-
nvidia/NVIDIA-Nemotron-Nano-12B-v2
Text Generation • 12B • Updated • 37.8k • • 145 -
nvidia/NVIDIA-Nemotron-Nano-9B-v2
Text Generation • 9B • Updated • 74.5k • 452 -
nvidia/NVIDIA-Nemotron-Nano-9B-v2-Base
Text Generation • 9B • Updated • 120k • 42 -
nvidia/NVIDIA-Nemotron-Nano-12B-v2-Base
Text Generation • 12B • Updated • 3.81k • 86
-
nvidia/llama-embed-nemotron-8b
Feature Extraction • 8B • Updated • 62.7k • 92 -
nvidia/omni-embed-nemotron-3b
Feature Extraction • 5B • Updated • 1.56k • 76 -
nvidia/llama-nemoretriever-colembed-3b-v1
Visual Document Retrieval • 4B • Updated • 612 • 65 -
nvidia/llama-nemoretriever-colembed-1b-v1
Visual Document Retrieval • 2B • Updated • 2.74k • 18
Collection of models for OpenReasoning-Nemotron which are trained on 5M reasoning traces for Math, Code and Science.
-
nvidia/OpenReasoning-Nemotron-1.5B
Text Generation • 2B • Updated • 479 • 49 -
nvidia/OpenReasoning-Nemotron-7B
Text Generation • 8B • Updated • 641 • • 47 -
nvidia/OpenReasoning-Nemotron-14B
Text Generation • 15B • Updated • 1.42k • 42 -
nvidia/OpenReasoning-Nemotron-32B
Text Generation • 33B • Updated • 1.07k • • 119
Mamba-Transformer hybrid models
-
nvidia/Nemotron-H-47B-Reasoning-128K
Text Generation • 47B • Updated • 681 • 18 -
nvidia/Nemotron-H-8B-Reasoning-128K
Text Generation • 8B • Updated • 3k • 24 -
nvidia/Nemotron-H-8B-Reasoning-128K-FP8
Text Generation • 8B • Updated • 66 • 12 -
nvidia/Nemotron-H-47B-Reasoning-128K-FP8
Text Generation • 47B • Updated • 46 • 5
Nemotron-4: open models for Synthetic Data Generation (SDG). Includes Base, Instruct, and Reward models.
Large scale pre-training datasets used in the Nemotron family of models.
Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
Large scale pre-training datasets used in the Nemotron family of models.
Open, Production-ready Enterprise Models
-
nvidia/Llama-3_3-Nemotron-Super-49B-v1_5
Text Generation • 50B • Updated • 391k • 216 -
nvidia/Llama-3_3-Nemotron-Super-49B-v1_5-FP8
Text Generation • 50B • Updated • 2.63k • 23 -
nvidia/Llama-3_1-Nemotron-Ultra-253B-v1
Text Generation • 253B • Updated • 170k • • 341 -
nvidia/Llama-3_3-Nemotron-Super-49B-v1
Text Generation • 50B • Updated • 14.2k • 320
A collection of multilingual, region-specific synthetic persona datasets that support sovereign AI development across many countries and regions.
Nemotron reward models. For use in RLHF pipelines and LLM-as-a-Judge
-
nvidia/Llama-3_3-Nemotron-Super-49B-GenRM
Text Generation • 50B • Updated • 151 • 17 -
nvidia/Llama-3_3-Nemotron-Super-49B-GenRM-Multilingual
Text Generation • 50B • Updated • 138 • 6 -
nvidia/Llama-3.3-Nemotron-70B-Reward
Text Generation • 71B • Updated • 552 • 2 -
nvidia/Llama-3.3-Nemotron-70B-Reward-Multilingual
Text Generation • 71B • Updated • 103 • 10
Novel ITS approach for open-ended tasks - No. 1 on Arena Hard on 18 Mar 2025
SOTA models on Arena Hard and RewardBench as of 1 Oct 2024.
-
nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
Text Generation • 71B • Updated • 6.41k • • 2.06k -
nvidia/Llama-3.1-Nemotron-70B-Reward-HF
71B • Updated • 1.7k • 89 -
nvidia/HelpSteer2
Viewer • Updated • 21.4k • 5.11k • 434 -
HelpSteer2-Preference: Complementing Ratings with Preferences
Paper • 2410.01257 • Published • 24
The Nemotron 3 8B Family of models is optimized for building production-ready generative AI applications for the enterprise.
Open, Production-ready Enterprise Models
-
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16
Text Generation • 32B • Updated • 315 • 49 -
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
Text Generation • 32B • Updated • 10.5k • 250 -
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8
Text Generation • 32B • Updated • 12.7k • 94 -
nvidia/Qwen3-Nemotron-235B-A22B-GenRM
Text Generation • 235B • Updated • 48 • 6
Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
Collection of datasets used in the post-training phase of Nemotron Nano v3.
Large scale pre-training datasets used in the Nemotron family of models.
Open, Production-ready Enterprise Models. Nvidia Open Model license.
-
nvidia/NVIDIA-Nemotron-Nano-12B-v2
Text Generation • 12B • Updated • 37.8k • • 145 -
nvidia/NVIDIA-Nemotron-Nano-9B-v2
Text Generation • 9B • Updated • 74.5k • 452 -
nvidia/NVIDIA-Nemotron-Nano-9B-v2-Base
Text Generation • 9B • Updated • 120k • 42 -
nvidia/NVIDIA-Nemotron-Nano-12B-v2-Base
Text Generation • 12B • Updated • 3.81k • 86
Open, Production-ready Enterprise Models
-
nvidia/Llama-3_3-Nemotron-Super-49B-v1_5
Text Generation • 50B • Updated • 391k • 216 -
nvidia/Llama-3_3-Nemotron-Super-49B-v1_5-FP8
Text Generation • 50B • Updated • 2.63k • 23 -
nvidia/Llama-3_1-Nemotron-Ultra-253B-v1
Text Generation • 253B • Updated • 170k • • 341 -
nvidia/Llama-3_3-Nemotron-Super-49B-v1
Text Generation • 50B • Updated • 14.2k • 320
-
nvidia/llama-embed-nemotron-8b
Feature Extraction • 8B • Updated • 62.7k • 92 -
nvidia/omni-embed-nemotron-3b
Feature Extraction • 5B • Updated • 1.56k • 76 -
nvidia/llama-nemoretriever-colembed-3b-v1
Visual Document Retrieval • 4B • Updated • 612 • 65 -
nvidia/llama-nemoretriever-colembed-1b-v1
Visual Document Retrieval • 2B • Updated • 2.74k • 18
A collection of multilingual, region-specific synthetic persona datasets that support sovereign AI development across many countries and regions.
Collection of models for OpenReasoning-Nemotron which are trained on 5M reasoning traces for Math, Code and Science.
-
nvidia/OpenReasoning-Nemotron-1.5B
Text Generation • 2B • Updated • 479 • 49 -
nvidia/OpenReasoning-Nemotron-7B
Text Generation • 8B • Updated • 641 • • 47 -
nvidia/OpenReasoning-Nemotron-14B
Text Generation • 15B • Updated • 1.42k • 42 -
nvidia/OpenReasoning-Nemotron-32B
Text Generation • 33B • Updated • 1.07k • • 119
Nemotron reward models. For use in RLHF pipelines and LLM-as-a-Judge
-
nvidia/Llama-3_3-Nemotron-Super-49B-GenRM
Text Generation • 50B • Updated • 151 • 17 -
nvidia/Llama-3_3-Nemotron-Super-49B-GenRM-Multilingual
Text Generation • 50B • Updated • 138 • 6 -
nvidia/Llama-3.3-Nemotron-70B-Reward
Text Generation • 71B • Updated • 552 • 2 -
nvidia/Llama-3.3-Nemotron-70B-Reward-Multilingual
Text Generation • 71B • Updated • 103 • 10
Mamba-Transformer hybrid models
-
nvidia/Nemotron-H-47B-Reasoning-128K
Text Generation • 47B • Updated • 681 • 18 -
nvidia/Nemotron-H-8B-Reasoning-128K
Text Generation • 8B • Updated • 3k • 24 -
nvidia/Nemotron-H-8B-Reasoning-128K-FP8
Text Generation • 8B • Updated • 66 • 12 -
nvidia/Nemotron-H-47B-Reasoning-128K-FP8
Text Generation • 47B • Updated • 46 • 5
Novel ITS approach for open-ended tasks - No. 1 on Arena Hard on 18 Mar 2025
SOTA models on Arena Hard and RewardBench as of 1 Oct 2024.
-
nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
Text Generation • 71B • Updated • 6.41k • • 2.06k -
nvidia/Llama-3.1-Nemotron-70B-Reward-HF
71B • Updated • 1.7k • 89 -
nvidia/HelpSteer2
Viewer • Updated • 21.4k • 5.11k • 434 -
HelpSteer2-Preference: Complementing Ratings with Preferences
Paper • 2410.01257 • Published • 24
Nemotron-4: open models for Synthetic Data Generation (SDG). Includes Base, Instruct, and Reward models.
The Nemotron 3 8B Family of models is optimized for building production-ready generative AI applications for the enterprise.
Large scale pre-training datasets used in the Nemotron family of models.