ChronoEdit-14B is an image editing and world simulation model with temporal reasoning capabilities developed by NVIDIA, with 14 billion parameters. It achieves physically-aware image editing and action-conditioned world simulation through a two-stage reasoning process, extracting prior knowledge from pre-trained video generation models.

MultimodalDiffusersMultiple Languages

NVIDIA Nemotron Parse v1.1 TC is an advanced document semantic understanding model that can extract text and table elements with spatial positioning from images and generate structured annotations, including formatted text, bounding boxes, and semantic categories. Compared with the previous version, the speed is increased by 20%, and the page order of unordered elements is retained.

MultimodalTransformers

NVIDIA Nemotron Parse v1.1 is an advanced document parsing model specifically designed to understand document semantics and extract text and table elements with spatial positioning. It can convert unstructured documents into machine-readable structured representations, overcoming the limitations of traditional OCR in handling complex document layouts.

MultimodalTransformers

NV-Reason-CXR-3B GGUF is a quantized version of the NVIDIA NV-Reason-CXR-3B vision-language model, optimized for edge device deployment. This is a model with 3 billion parameters, focusing on chest X-ray analysis. It has been converted to the GGUF format and quantized for efficient operation on mobile devices, desktops, and embedded systems.

MultimodalGgufEnglish

This is a 32B parameter reward model developed by NVIDIA based on the Qwen3 architecture. It is specifically used for reward scoring and principle alignment in reinforcement learning, helping to train AI systems that are safer and more in line with human values.

Natural Language ProcessingGguf

This is the GGUF quantized version of NVIDIA's Qwen3-Nemotron-32B-RLBFF large language model. It uses the llama.cpp tool for multi-precision quantization, offering more than 20 quantization options from BF16 to IQ2_XXS, suitable for different hardware configurations and performance requirements.

Natural Language ProcessingGguf

This is the GGUF quantized version of the NVIDIA ChronoEdit-14B-Diffusers model, specifically designed for image-to-video tasks. This model retains all the functions of the original model while optimizing deployment and runtime efficiency through the GGUF format.

Computer VisionGguf

Qwen3-VL-2B-Thinking is one of the most powerful vision-language models in the Qwen series. It uses GGUF format weights and supports efficient inference on devices such as CPUs, NVIDIA GPUs, and Apple Silicon. This model has excellent multimodal understanding and reasoning capabilities, especially enhancing visual perception, spatial understanding, and agent interaction functions.

MultimodalTransformers

TheWhisper-Large-V3 is a high-performance fine-tuned version of the OpenAI Whisper Large V3 model, optimized by TheStage AI for real-time, low-latency, and low-power consumption speech-to-text inference on multiple platforms (NVIDIA GPU and Apple Silicon).

Audio ProcessingTransformersOther

NVIDIA-Nemotron-Nano-VL-12B-V2-FP4-QAD is an autoregressive vision-language model launched by NVIDIA. Based on an optimized Transformer architecture, it can handle both image and text inputs simultaneously. The model uses FP4 quantization technology, which significantly reduces the model size and inference cost while maintaining performance, and is suitable for various multimodal application scenarios.

MultimodalTransformers

NVIDIA-Nemotron-Nano-VL-12B-V2-FP8 is a quantized vision-language model launched by NVIDIA, which adopts an optimized Transformer architecture and has undergone three-stage training on commercial images. This model supports single-image inference, has multilingual and multimodal processing capabilities, and is suitable for various scenarios such as image summarization and text-image analysis.

MultimodalTransformers

BR-RM is an innovative two-round reasoning reward model that solves the 'judgment diffusion' problem in traditional reward models through adaptive branching and branch-based reflection mechanisms, achieving industry-leading performance in multiple reward modeling benchmark tests.

Natural Language ProcessingTransformersEnglish

NVIDIA Nemotron Nano v2 12B VL is a powerful multimodal vision-language model that supports multi-image reasoning and video understanding, and has document intelligence, visual question answering, and summarization capabilities. It can be used for commercial purposes.

MultimodalTransformers

This is a robot operation model fine-tuned by Tacoin on the LIBERO libero long benchmark based on the NVIDIA GR00T model. The model uses dual RGB streams and 8-degree-of-freedom state input, and can predict 16-step joint space actions, specifically designed for long-horizon robot operation tasks.

MultimodalSafetensors

Llama Nemotron Reranking 1B is a model developed by NVIDIA specifically for text retrieval reordering. It is fine-tuned based on the Llama-3.2-1B architecture and can provide a relevance log score for query-document pairs. It supports multilingual and long document processing.

Natural Language ProcessingTransformersOther

The Llama Nemotron Embedding 1B model is an embedding model developed by NVIDIA, optimized for multilingual and cross - language text question - answering retrieval. It supports 26 languages, can handle documents up to 8192 tokens long, and can significantly reduce data storage requirements through dynamic embedding sizes.

Natural Language ProcessingTransformersOther

Nemotron-Flash-3B is a new hybrid small language model launched by NVIDIA, specifically designed for low-latency requirements in practical applications. This model demonstrates excellent performance in tasks such as mathematics, coding, and common-sense reasoning, and also has the characteristics of excellent low latency for small batches and high throughput for large batches.

Natural Language ProcessingTransformers

Qwen3-Nemotron-32B-RLBFF is a large language model fine-tuned based on Qwen/Qwen3-32B. The quality of the model's generated responses in the default thinking mode has been significantly improved through reinforcement learning feedback technology. This model performs excellently in multiple benchmark tests while maintaining low inference costs.

Natural Language ProcessingTransformersEnglish

NVIDIA GPT-OSS-120B Eagle3 is an optimized version based on the OpenAI gpt-oss-120b model. It adopts the Mixture of Experts (MoE) architecture, with a total of 120 billion parameters and 5 billion active parameters. This model supports both commercial and non-commercial use and is suitable for text generation tasks, especially for the development of AI Agent systems, chatbots, and other applications.

Natural Language ProcessingSafetensors

This is the FP8 dynamic quantization version of the NVIDIA-Nemotron-Nano-9B-v2 model. Optimization is achieved by quantizing weights and activations to the FP8 data type, significantly reducing disk size and GPU memory requirements by approximately 50% while maintaining excellent text generation performance.

Natural Language ProcessingTransformersMultiple Languages