NVIDIA · Embedding
Llama Nemotron Embed VL 1b V2
Nemotron Embed VL 1B v2 is NVIDIA's open-weight embedding model from its Llama Nemotron family, trained for retrieval over visually rich documents such as reports, slides and scanned pages. Through the API it embeds text into 2,048-dimensional vectors.
Evidence and sources
🤗 nvidia/llama-nemotron-embed-vl-1b-v2llminfo IQ ≈ 116, ranked among embedding models. From 5 benchmarks. Little data behind it, so treat it as a rough estimate.How llminfo IQ is computedReleased 18 Dec 2025Maker HQ: US (source)
All rankings:
#432 on OpenRouter this month
#456 on OpenRouter this weekSources: DeepInfra data, Venice.ai data, Models.dev data, llminfo researchPrimary data from DeepInfra
from Venice.ai: primary mode, description
from Models.dev: weights availability
from llminfo research: release date
Input & output
as stated by the model's providersText
Embeddings
Also known as: nemotron embed vl 1b v2, nemotron-embed-vl-1b-v2, nvidia/llama-nemotron-embed-vl-1b-v2, nvidia_nim/nvidia/llama-nemotron-embed-vl-1b-v2
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works