Nemotron 3 Embed 1B BF16

Jul 2026
Nemotron 3 Embed 1B BF16
Compare in
on
$0.015per 1M tokens
56.8RTEB French beta
Jul 2026released
2,048Dimensions
33K tokensMax input
en, es, fr +30Languages

Ranking?

Full ranking →
1Octen Law 8B v176.82MoD Embedding74.63Litil Embed 0.6B73.34Dinghy Law 8B v172.65Nemotron 3 Embed 8B BF1671.8
···
9Nemotron 3 Embed 8B Legal69.310Nemotron 3 Embed 1B BF1668.611Dinghy Law 0.6B v165.8

Alternatives?

ModelScoreCostPriceWhat changes
lBetterOcten Law 8B v1Open+12%——Languages: loses es, fr, ja +27+1bBetterMoD EmbeddingOpen+9%——Languages: loses es, fr, de +28+1lBetterLitil Embed 0.6BOpen+7%——Languages: loses es, fr, ja +27HBetterDinghy Law 8B v1Open+6%——Max input: 33K tokens → 8K tokens+2BetterNemotron 3 Embed 8B BF16Open+5%——Memory (4-bit): 0.7 GB → 4.8 GB

Run it yourself

estimate
Fits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licenceopenmdw-1.1commercial use with conditions

Benchmarks

in Embeddings & Retrieval
MTEB German v1 · Mean over tasks · German38.1#76
MTEB German v1 · Mean over task types · German41.3#53
MTEB German v1 · Classification · German37.9#163
MTEB German v1 · Pair Classification · German51.1#73
MTEB German v1 · STS · German37.2#128
MTEB German v1 · Clustering · German7.1#162
MTEB German v1 · Retrieval · German59.1#3
MTEB German v1 · Reranking · German55.4#2
MTEB Dutch v1 · Mean over tasks · Dutch40.9#77
MTEB Dutch v1 · Mean over task types · Dutch46.4#92
MTEB Dutch v1 · Retrieval · Dutch61.2#21
MTEB Dutch v1 · Classification · Dutch39.1#126
MTEB Dutch v1 · Multilabel Classification · Dutch20.8#131
MTEB Dutch v1 · STS · Dutch45.8#129
MTEB Dutch v1 · Clustering · Dutch8.7#133
MTEB Dutch v1 · Reranking · Dutch90.4#30
MTEB Dutch v1 · Pair Classification · Dutch58.9#114
MTEB Law v1 · Mean over tasks · multilingual (3 languages)68.6#10
MTEB Law v1 · Mean over task types · multilingual (3 languages)68.6#10
MTEB Law v1 · Retrieval · multilingual (3 languages)68.6#10
RTEB Finnish beta · Mean over tasks · English77.4#18
RTEB French beta · Mean over tasks · multilingual (2 languages)56.8#20
RTEB German beta · Mean over tasks · German73.1#12
RTEB Health beta · Mean over tasks · multilingual (4 languages)63.1#23
RTEB Law beta · Mean over tasks · multilingual (4 languages)61.5#20

FAQ

How much does Nemotron 3 Embed 1B BF16 cost?›

$0.015 per 1M tokens.

Is Nemotron 3 Embed 1B BF16 free to use?›

There is no free API; the cheapest paid price is $0.015 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Nemotron 3 Embed 1B BF16, and what is it best at?›

#20 of 215 in Embeddings.

Is Nemotron 3 Embed 1B BF16 better than Nemotron 3 Embed 8B Legal?›

On MTEB Law v1 · Mean over task types · multilingual (3 languages), Nemotron 3 Embed 1B BF16 scores 68.6 vs 69.3 — about the same.

How can I use Nemotron 3 Embed 1B BF16?›

Through the NVIDIA API.

When was Nemotron 3 Embed 1B BF16 released? Is there a newer version?›

Released 2026-07-16. It is the current version in its line.

Is Nemotron 3 Embed 1B BF16 open source?›

Yes, the weights are open (openmdw-1.1 licence).

Can I run Nemotron 3 Embed 1B BF16 locally?›

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

What is Nemotron 3 Embed 1B BF16?›

Nemotron 3 Embed 1B BF16 is an open-weights embeddings & retrieval model from NVIDIA, released Jul 2026.

What are the best alternatives to Nemotron 3 Embed 1B BF16?›

Octen Law 8B v1 (better).