SigLIP 2

Model Overview

Description:

SigLIP 2 is a vision-language encoder that produces joint image and text embeddings for cross-modal retrieval. It supports multiple variants (e.g., SO400M, giant) with embedding dimensions from 768 to 1536 and image resolutions such as 224×224, 256×256, 384×384, and 512×512 depending on variant. The model is suitable for text-to-image and image-to-image search, object search, and re-identification in applications such as transportation, warehouse operations, and other industrial contexts. This model is ready for commercial use.

License/Terms of Use

Governing Terms: Use of this model is governed by the NVIDIA Open Model License. Additional Information: Apache 2.0.

Deployment Geography:

Global

Use Case:

Object search, re-identification, and cross-modal retrieval (text-to-image and image-to-image) in applications such as transportation, warehouse operations, and other industrial contexts. Users include developers and integrators building vision-language retrieval systems for these domains.

Release Date:

NGC 03/09/2026 via URL

References(s):

SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features, 2025.

Model Architecture:

Architecture Type: Vision-Language Transformer (SigLIP 2)

Network Architecture: SigLIP 2 (e.g., siglip2-so400m-patch16-256, siglip2-so400m-patch16-384, siglip2-giant-opt-patch16-384, or variant as specified on NGC).

This model was developed based on: Google SigLIP 2 (vision and text encoders).

Number of model parameters: Varies by variant (e.g., ~4×10^8 for SO400M, ~1×10^9 for giant).

Input(s):

Input Type(s): Image, Text

Input Format(s):

• Image: Red, Green, Blue (RGB)
  
• Text: String (tokenized via model processor/tokenizer)
  

Input Parameters:

• Image: Two-Dimensional (2D)
  
• Text: One-Dimensional (1D)
  

Other Properties Related to Input: Image: Resolution depends on variant (e.g., 256×256, 384×384, 512×512); RGB, 8-bit; pre-processing: resize and normalize per HuggingFace processor. Some variants support dynamic resolution (NaFlex). Text: Context length and vocabulary per model processor; pre-processing: tokenization.

Output(s)

Output Type(s): Embedding vectors (floating-point)

Output Format(s): Real-valued vectors (L2-normalized)

Output Parameters: One-Dimensional (1D) embedding vectors

Other Properties Related to Output: Embedding dimension varies by variant (e.g., 1152 for SO400M, 1536 for giant); L2-normalized for cosine similarity; no character or resolution limits.

Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.

Software Integration:

Runtime Engine(s):

• TAO - minimally compatible versions as specified on NGC
  
• PyTorch / HuggingFace Transformers
  

Supported Hardware Microarchitecture Compatibility:

• NVIDIA Ampere
  
• NVIDIA Blackwell
  
• NVIDIA Jetson
  
• NVIDIA Hopper
  
• NVIDIA Lovelace
  
• NVIDIA Pascal
  
• NVIDIA Turing
  
• NVIDIA Volta
  

Preferred/Supported Operating System(s):

• Linux
  

The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.

This AI model can be embedded as an Application Programming Interface (API) call into the software environment described above.

Model Version(s):

Trainable v1.0 and deployable v1.0 are available on NGC (e.g., SigLIP 2 SO400M).

Training, Testing, and Evaluation Datasets:

Training Dataset

Data Modality:

• Image
  
• Text
  

Image Training Data Size:

• Less than a Million Images (for fine-tuning workflows; base model training may use larger sets). TAO fine-tuning training set sizes: 50,019 train images (6,668 persons).
  

Text Training Data Size:

• Less than a Billion Tokens
  

Data Collection Method by dataset:

• Hybrid: Human, Automated
  

Labeling Method by dataset:

• Hybrid: Human, Automated
  

Properties: Training data comprises image-text pairs: images (RGB) and associated text (captions or attribute-based descriptions). Content includes person and object imagery for retrieval and re-identification. Data sources used in TAO fine-tuning may include internal and licensed datasets (e.g., person attribute search); see the CLIP CLI notebook and experiment specs for exact sources and counts.

Testing Dataset:

Data Collection Method by dataset:

• Hybrid: Human, Automated
  

Labeling Method by dataset:

• Human
  

Properties: Test sets include image-text pairs with same modalities and nature as training (images and captions/attributes); used for retrieval and re-identification metrics.

Evaluation Dataset:

Data Collection Method by dataset:

• Human
  

Labeling Method by dataset:

• Human
  

Properties: Evaluation uses benchmark datasets for retrieval and person re-identification (e.g., person attribute search benchmarks); image and text modalities; descriptive and attribute-based captions.

Benchmark Score: Retrieval metrics (e.g., Recall@K, mAP) on person re-identification and text-to-image retrieval benchmarks.

Inference

Acceleration Engine: TensorRT (SigLIP 2 vision encoder), PyTorch / Hugging Face Transformers

Test Hardware:

• NVIDIA datacenter GPUs (e.g., L4, L40, L40S, A100, H100, H200, RTX PRO 6000 Blackwell Server Edition, B200, GB200)
  
• Jetson AGX Orin, Jetson AGX Thor T5000, Jetson IGX Thor T7000 (Stargazer), DGX Spark
  
• Minimum hardware as required by your TAO / PyTorch / TensorRT deployment
  

Datacenter GPU results below are for the SigLIP 2 SO400M patch16-256 vision encoder (google/siglip2-so400m-patch16-256). Measurements used TensorRT FP16 and trtexec. Throughput is inference-only; end-to-end latency with decoding, pre/post-processing, or full application pipelines may differ.

Environment (dGPU)

Environment (edge)

Jetson IGX Thor T7000 (Stargazer) — flashed with IGX-r38v2.0.11; TensorRT 10.13.3.9; CUDA 13.0; cuDNN 9.12.0.46; power mode 120W; jetson_clocks applied.

Jetson AGX Thor T5000 — flashed with JP7.1 b148; TensorRT 10.13.3.9; CUDA 13.0; cuDNN 9.12.0.46; power mode 120W; jetson_clocks applied.

DGX Spark — FastOS OTA2 Mainline Release 1.120.36; driver 580.126.09 / 590.48.01; CUDA 13.0 / 13.1; cuDNN 9.12 / 9.17; TensorRT 10.13.2 / 10.14.1.48.

Jetson AGX Orin — flashed with JP7.2 b19; TensorRT 10.13.3.9; CUDA 12.9; cuDNN 9.12.0.46; power mode MAXN.

Model configuration and TensorRT build

Hugging Face model id: google/siglip2-so400m-patch16-256. Input: pixel_values shape [batch_size, 3, 256, 256]. Output: image_embeds shape [batch_size, 1152]. ONNX opset 17.

TensorRT conversion command (from export log):

trtexec --onnx=onnx_exports/siglip2_v1.0.onnx --saveEngine=siglip2_v1.0.engine --shapes=pixel_values:1x3x256x256 --fp16

Vision encoder throughput (FP16)

Ethical Considerations:

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.

Please make sure you have proper rights and permissions for all input image and video content; if image or video includes people, personal health information, or intellectual property, the image or video generated will not blur or maintain proportions of image subjects included.

For more detailed information on ethical considerations for this model, please see the Model Card++ Bias, Explainability, Safety & Security, and Privacy Subcards.

Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns here.