Nemotron-3.5-Embed-8B-BF16

Name: Nemotron-3.5-Embed-8B-BF16
Author: DeepInfra

embed

Nemotron-3.5-Embed-8B-BF16 is a multimodal embedding model from NVIDIA that maps text, images, and page screenshots into a shared 4096-dimensional vector space for visual document retrieval and semantic search. Built on the Nemotron-3-Embed-8B text encoder with a vision encoder, it supports up to 32k tokens and Matryoshka dimensions, and delivers the highest retrieval quality in the Nemotron 3.5 family.

Capabilities

Show unsupported (7)

Infrastructure

Streaming

Details

Model IDnvidia/Nemotron-3.5-Embed-8B-BF16

ProviderDeepInfra

CreatorNVIDIA

Family—

License—

Statusactive

Typeembed

Input modalitiestext, image

Output modalitiesembedding

Open weightYes

Parameters—

Architecture—

Context window—

Max input—

Max output—

Reasoning tokens—

Knowledge cutoff—

Release date2026-10-20

Sourceofficial

Endpoints

EmbeddingsPOST

Generate vector embeddings for texthttps://api.deepinfra.com/v1/openai/v1/embeddings

API

GET/v1/models/deepinfra/nvidia/Nemotron-3.5-Embed-8B-BF16