> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/vss/dev/system-components/models/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/vss/_mcp/server.

# Models

Documentation for **fine-tuning, exporting, and integrating** models used with the VSS blueprint: **embedding** models for semantic search (RT-CV and RT-Embedding), and **computer vision** models for warehouse perception (TAO Sparse4D and RT-DETR).

Start with [Model customization overview](/system-components/models/model-customization-overview) for a cross-service summary, then dive into the per-model pages below.

## Docs

- [Model customization overview](https://docs.nvidia.com/vss/dev/system-components/models/model-customization-overview.md): 
- [SigLIP 2 object embeddings](https://docs.nvidia.com/vss/dev/system-components/models/siglip-2-object-embeddings.md): 
- [Cosmos-Embed1 (video embedding)](https://docs.nvidia.com/vss/dev/system-components/models/cosmos-embed1-video-embedding.md): 
- [Sparse4D (3D multi-camera)](https://docs.nvidia.com/vss/dev/system-components/models/sparse4d-3d-multi-camera.md): 
- [RT-DETR (2D warehouse detection)](https://docs.nvidia.com/vss/dev/system-components/models/rt-detr-2d-warehouse-detection.md):