nvidia

Collection of logits processors to customize LLM generation for specific tasks. Drop-in for vLLM, HF, TRT-LLM.

Text normalization and inverse text normalization toolkit for ASR and TTS pipelines.

Lightweight C++ LLM and VLM inference engine optimized for edge devices and physical AI.

NVIDIA GEAR Lab's DreamGen initiative tackling the robotics data problem using world models for embodied AI training.

Open-source deep-learning framework for building and deploying AI weather and climate prediction workflows.

NVIDIA's neural vocoder that synthesizes high-fidelity audio (speech, music, singing) from mel-spectrograms.

KV-cache compression toolkit for LLMs — drop-in techniques to cut memory and extend context length.

NVIDIA's foundation model for generalist humanoid robots (GR00T N1.7) — perception-to-action policy that transfers across body shapes and tasks.

NVIDIA's Python API for defining LLMs and running them on NVIDIA GPUs with state-of-the-art inference optimizations.

NVIDIA's LLM vulnerability scanner — probes models for jailbreaks, prompt injection, data leaks, hallucinations, and toxicity at scale.

NVIDIA's SDK for high-performance deep learning inference on NVIDIA GPUs — optimizes trained models for low-latency production deployment.

A real-time, full-duplex speech-to-speech conversational AI model that enables persona control through text-based role prompts and audio-based voice conditioning. Trained on synthetic and real conversations, it produces natural, low-latency spoken interactions with consistent personas.