Community · Oct 23, 2026, 04:46 UTC

Nvidia updates Nemotron embed and rerank models to version 3.5

Nvidia's Nemotron-3.5 family includes Parse, Embed in 1B and 8B BF16 sizes, and Rerank 8B BF16, per Hugging Face links in a Reddit post. Posts reports the models only through links and a user question about trying them.

• LabNVIDIA
• ModelNemotron-3.5-Parse
• ModelNemotron-3.5-Embed-1B-BF16
• ModelNemotron-3.5-Embed-8B-BF16
• ModelNemotron-3.5-Rerank-8B-BF16
• Topicrelease
• Topicresearch