Cosmos3-Edge
An edge visual encoding model in the NVIDIA Cosmos series that encodes images into visual tokens for use by a language decoder.
Timeline (last 90 days)
Industrial ReMEmbR uses Cosmos-Embed1 to generate embeddings for retrieval, Cosmos3-Edge to generate visual tokens, and Wan2.2 VAE to generate reconstruction latents.
Industrial ReMEmbR splits Cosmos3-Edge along the vision-language boundary: the vision encoder stays on the robot, and the language decoder is placed on a local server.
All Facts (2)
Industrial ReMEmbR uses Cosmos-Embed1 to generate embeddings for retrieval, Cosmos3-Edge to generate visual tokens, and Wan2.2 VAE to generate reconstruction latents.
50%UnverifiedIndustrial ReMEmbR splits Cosmos3-Edge along the vision-language boundary: the vision encoder stays on the robot, and the language decoder is placed on a local server.
50%