> This page is for version 1.1.0 (Latest) (default).
> For other versions, use one of these documentation indexes:
> - 1.1.0 (Latest) (default): https://docs.nvidia.com/nim/cosmos-embed1/1.1.0/llms.txt
> - 1.0.0: https://docs.nvidia.com/nim/cosmos-embed1/1.0.0/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nim/cosmos-embed1/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nim/cosmos-embed1/_mcp/server.

# Release Notes

## January 27, 2026

## Release 1.1.0

### Features

* Added `video_frames` input support (8 frames per item) for `query` and `bulk_video` modes.
* Added a new JSON health metrics endpoint: `GET /health/metrics`
* Updated the video decoding dependency to `PyNvVideoCodec==2.0.3`

## September 15, 2025

## Release 1.0.0

### Features

* First release of the Cosmos Embed1 NIM (224p variant; 256-dimensional embeddings)
* OpenAI Embeddings-compatible endpoint: `/v1/embeddings` supporting the following `request_type` values: `query`, `bulk_text`, and `bulk_video`
* Health endpoints: `/v1/health/ready` and `/v1/health/live`
* Standard container deployment; optional host-side model caching via `LOCAL_NIM_CACHE` mounted to `/opt/nim/.cache`.