NV Embed v2

Sep 2024
NV Embed v2
Compare in
on
self-host onlyprice
62.8MTEB English v2
Sep 2024released
4,096Dimensions
33K tokensMax input
enLanguages

Run it yourself

estimate
Fits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceCC BY-NC 4.0no commercial use

Benchmarks

in Embeddings & Retrieval

FAQ

How much does NV Embed v2 cost?›

There is no per-unit API price: self-host only.

Is NV Embed v2 free to use?›

There is no free API. The weights are open, so you can run it yourself.

How good is NV Embed v2, and what is it best at?›

#12 of 200 in Embeddings. Strongest at: Embeddings & Retrieval.

Is NV Embed v2 better than BOOM 4B v1?›

On MTEB English v2 · Classification · English, NV Embed v2 scores 87.2 vs 87.5 — about the same.

When was NV Embed v2 released? Is there a newer version?›

Released 2024-09-09. It is the current version in its line.

Is NV Embed v2 open source?›

Yes, the weights are open (cc-by-nc-4.0 licence).

Can I run NV Embed v2 locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

What is NV Embed v2?›

NV Embed v2 is an open-weights embeddings & retrieval model from NVIDIA, released Sep 2024.

What are the best alternatives to NV Embed v2?›

Qwen 3 Embedding 8B (better), F2LLM v2 14B (open weights).