Tag
Explore by tags
Argues a single web-scale generative video model handles vision tasks zero-shot the way LLMs handle language. Probes Veo 3 on segmentation, edge detection, image editing, physical and affordance reasoning, and puzzles like maze solving and symmetry.
Provides a DiT-based audio–video foundation model plus an official Python inference and LoRA trainer. Ships multiple production-ready pipelines (text/image/audio→video), checkpoints, and performance optimizations (FP8, distilled pipelines) for high-fidelity synchronized audio–video generation.
Provides a conditional memory module that performs O(1) N‑gram lookups and fuses static embeddings into transformer hidden states — enables offloading large embedding tables to host memory with minimal inference overhead.
A 26M-parameter LLM distilled for reliable function-call generation on tiny devices, with open weights, local finetuning tooling, and a web playground for on-device testing. Pretrained at scale then post-trained on a single-shot function-call dataset for tool integration.
Paired brain MRI scans and radiology text annotations for multimodal vision–language research. Provides image-level labels and image–text pairs suited for VQA, classification, and image-to-text tasks; CC BY-NC-SA 4.0 and ~10K–100K samples — research/non-commercial use.
Generate text, images, video, audio and action/robot trajectories from combined text, image, video, audio and action inputs. A Mixture-of-Transformers omnimodal foundation model (Cosmos3‑Nano, 16B params) focused on Physical AI (robotics, AV, simulation) and optimized for NVIDIA GPU runtimes.
Instruction-tuned Gemma 4 31B multimodal model that generates text from text+image inputs with up to 256K context. Dense 31B variant optimized for vision-language understanding, long-context reasoning, and coding; Apache‑2.0 licensed.
Instruction-tuned Mixture-of-Experts multimodal model that generates text from text+image inputs while activating a 4B subset of parameters for faster inference; supports a 256K context window, multilingual vision-language tasks, and is available under Apache-2.0.
A dense 128B multimodal model with a 256k context window, configurable reasoning effort, and native function-calling for agentic workflows. Supports text+image input, multilingual output, and is released on Hugging Face under a Modified MIT license with revenue-based exceptions.
Generates and iterates on long‑horizon agentic plans and code — designed to stay productive across many rounds of tool calls and experiments. Emphasizes iterative reasoning, stronger repo/terminal automation and code generation than GLM‑5, and can be served locally for research and autonomous-agent workloads.
Generates persistent, explorable 3D worlds from a single image by synthesizing long-range, geometry-consistent video and reconstructing it into an explicit 3D Gaussian scene. Intended for internal research use under NVIDIA's research license.
A 30B-parameter, instruction-tuned language model built for long-context text generation, conversational agents, and tool-calling. It combines supervised fine-tuning and RL alignment, supports 131,072-token context, and is optimized for tasks like summarization, code, and RAG.