ReferenceAsk
Remy Reference/models/Nemotron 3.5 Embed 8B
D

Nemotron 3.5 Embed 8B

Nemotron 3.5 Embed 8B, an embedding model from DeepInfra.
model: "nemotron-3.5-embed-8b"
Nvidia·served viaDeepInfra·Released October 2026
Pricing
Per 1M tokens
$0.035 / 1M
Specifications
Context window
32,768 tokens
Dimensions
4,096
Max input
32,768 tokens
Max batch
96
Input types
document, query
Released
October 2026
Model string
nvidia/Nemotron-3.5-Embed-8B-BF16
Use this model
// Embedding models power the vector database. This model is
// selected when a corpus is configured, not called directly.
// Model id: 0