Converts streamed audio to facial blendshapes for realtime lipsyncing and facial performances.

State-of-the-art accuracy and speed for English transcriptions.

Generative downscaling model for generating high resolution regional scale weather fields.

FourCastNet predicts global atmospheric dynamics of various weather / climate variables.

Efficiently refine retrieval results over multiple sources and languages.

World-class multilingual and cross-lingual question-answering retrieval.

A bilingual Hindi-English SLM for on-device inference, tailored specifically for Hindi Language.

Shutterstock Generative 3D service for 360 HDRi generation. Trained on NVIDIA Edify using Shutterstock’s licensed creative libraries.

Llama-3.1-Nemotron-70B-Instruct is a large language model customized by NVIDIA in order to improve the helpfulness of LLM generated responses.

Enhance speech by correcting common audio degradations to create studio quality speech output.

State-of-the-art small language model delivering superior accuracy for chatbot, virtual assistants, and content generation.

Leaderboard topping reward model supporting RLHF for better alignment with human preferences.

Unique language model that delivers an unmatched accuracy-efficiency performance.

Generates consistent characters across a series of images without requiring additional training.

Multi-modal vision-language model that understands text/images and generates informative responses

Optimized SLM for on-device inference and fine-tuned for roleplay, RAG and function calling

State-of-the-art small language model delivering superior accuracy for chatbot, virtual assistants, and content generation.

NV-DINOv2 is a visual foundation model that generates vector embeddings for the input image.

Grounding dino is an open vocabulary zero-shot object detection model.

Natural, high-fidelity, English voices for personalizing text-to-speech services and voiceovers

Enable smooth global interactions in 32 languages.

Expressive and engaging English voices for Q&A assistants, brand ambassadors, and service robots

Record-setting accuracy and performance for English transcription.

State-of-the-art accuracy and speed for English transcriptions.

State-of-the-art LLM that answers OpenUSD knowledge queries and generates USD-Python code.

AI-powered search for OpenUSD data, 3D models, images, and assets using text or image-based inputs.

Shutterstock Generative 3D service for 3D asset generation. Trained on NVIDIA Edify using Shutterstock’s licensed creative libraries

Getty Images’ API service for 4K image generation. Trained on NVIDIA Edify using Getty Images' commercially safe creative libraries.

Estimate gaze angles of a person in a video and redirect to make it frontal.

Create facial animations using a portrait photo and synchronize mouth movement with audio.

Verify compatibility of OpenUSD assets with instant RTX render and rule-based validation.

GPU-accelerated model optimized for providing a probability score that a given passage contains the information to answer a question.

GPU-accelerated generation of text embeddings used for question-answering retrieval.

GPU-accelerated generation of text embeddings used for question-answering retrieval.

MAISI is a pre-trained volumetric (3D) CT Latent Diffusion Generative Model.

Advanced LLM to generate high-quality, context-aware responses for chatbots and search engines.

Advanced LLM to generate high-quality, context-aware responses for chatbots and search engines.

Grades responses on five attributes helpfulness, correctness, coherence, complexity and verbosity.

Creates diverse synthetic data that mimics the characteristics of real-world data.

NV-CLIP is a multimodal embeddings model for image and text.

OCDNet and OCRNet are pre-trained models designed for optical character detection and recognition respectively.

Generates high-quality numerical embeddings from text inputs.

Visual Changenet detects pixel-level change maps between two images and outputs a semantic change segmentation mask

EfficientDet-based object detection network to detect 100 specific retail objects from an input video.

GPU-accelerated generation of text embeddings used for question-answering retrieval.

GPU-accelerated model optimized for providing a probability score that a given passage contains the information to answer a question.

Multi-modal vision-language model that understands text/images and generates informative responses

VISTA-3D is a specialized interactive foundation model for segmenting and anotating human anatomies.

This LLM follows instructions, completes requests, and generates creative text.

Generate BAM output given one or more pairs of FASTQ files, by running BWA-MEM & GATK best practices.

Run Google's DeepVariant optimized for GPU. Switch models for high accuracy on all major sequencers.

MolMIM performs controlled generation, finding molecules with the right properties.

World-record accuracy and performance for complex route optimization.