Explore Vision Models | Try NVIDIA NIM APIs

Deploy Models Now with NVIDIA NIM

Optimized inference for the world’s leading models

Free serverless APIs for development

Self-Host on your GPU infrastructure

Continuous vulnerability fixes

Discover Models Blueprints GPUs

Docs

Forums

models

Reasoning Vision Visual Design Retrieval Speech Biology Simulation Climate & Weather Safety & Moderation

industries

Automotive Gaming Healthcare Industrial Robotics

Vision

Explore NVIDIA Blueprints

Comprehensive reference workflows that accelerate application development and deployment, featuring NVIDIA acceleration libraries, APIs, and microservices for AI agents, digital twins, and more.

nvidia Build a Video Search and Summarization (VSS) Agent

Ingest massive volumes of live or archived videos and extract insights for summarization and interactive Q&A

blueprint nvidia ai chat generative ai video-to-text vision

Vision Language Models (VLM)

Multimodal models that can reason against image and video inputs and perform descriptive language generation

Run Anywhere

meta llama-3.2-90b-vision-instruct

Cutting-edge vision-Language model exceling in high-quality reasoning from images.

image-text retrieval visual grounding visual qa image captioning image-to-text

Run Anywhere

meta llama-3.2-11b-vision-instruct

Cutting-edge vision-language model exceling in high-quality reasoning from images.

image captioning image-text retrieval visual grounding visual qa image-to-text

PREVIEW

nvidia vila

Multi-modal vision-language model that understands text/img/video and creates informative responses

vlm vision language model image caption image to text

PREVIEW

microsoft florence-2

Vision foundation model capable of performing diverse computer vision and vision language tasks.

language generation multimodal vision assistant visual question answering computer vision cv image image classification image-to-text object detection text-to-image vlm

PREVIEW

nvidia nv-dinov2

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

nvidia nim computer vision deepstream object classification image-to-embedding

PREVIEW

nvidia nv-grounding-dino

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

nvidia nim object detection computer vision deepstream object detection

PREVIEW

nvidia neva-22b

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

non-commercial use only vision assistant visual question answering computer vision cv image image-to-text video vlm

PREVIEW

microsoft phi-3-vision-128k-instruct

Cutting-edge open multimodal model exceling in high-quality reasoning from images.

language generation vision assistant visual question answering computer vision cv image image-to-text video vlm

PREVIEW

google paligemma

Vision language model adept at comprehending text and visual inputs to produce informative responses

language generation vision assistant visual question answering computer vision cv image image-to-text video vlm

PREVIEW

microsoft kosmos-2

Groundbreaking multimodal model designed to understand and reason about visual elements in images.

image understanding multimodal visual question answering computer vision cv image image-to-text video vlm

PREVIEW

adept fuyu-8b

Multi-modal model for a wide range of tasks, including image understanding and language generation.

image understanding language generation multimodal computer vision cv image image-to-text video vlm

Specialized Foundation Models

Computer vision models that excel at particular visual perception tasks

PREVIEW

meta sam2

SAM 2 is a segmentation model that enables fast, precise selection of any object in any video or image.

computer vision meta segmentation video

PREVIEW

nvidia ocdrnet

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

optical character detection optical character recognition tao toolkit computer vision cv image optical character recognition video vlm

PREVIEW

nvidia visual-changenet

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

image segmentation nvidia nim tao toolkit computer vision cv image image generation video vlm

Run Anywhere

nvidia nvclip

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

computer vision nvidia nim run on rtx multimodal embeddings text and image

PREVIEW

nvidia retail-object-detection

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

nvidia nim object detection tao toolkit computer vision cv image object detection video vlm