Skip to content
Local AI Wizard
Your machine
Theme

Nemotron 3 Embed 1B

NVIDIA · nvidia/Nemotron-3-Embed-1B-BF16

Weights license: OpenMDW 1.1 · code and components may use other terms

Reported by the publisher3 resultsNVIDIA's model card · read 2026-10-25 · not ranked · see them
Memory needed~4 GB8-bit planning estimate · not measured
Parameters1.14 BSee technical details for the basis
Dimensions · context2,04832,768 tokens

Quality

Not in the MTEB ranking

The MTEB results dataset covers 13 of 131 tasks of MTEB (Multilingual, v2); a mean over a partial task set is not comparable, so none is computed.

Reported by the publisher

3 results · NVIDIA's model card · read 2026-10-25
Text Retrieval benchmarks - Avg. NDCG@10 · RTEB (chunk retrieval)
72.38
Text Retrieval benchmarks - Avg. NDCG@10 · ViDoRe-V3 text (chunk retrieval)
57.74
Text Retrieval benchmarks - Avg. NDCG@10 · MMTEB (chunk retrieval; Retrieval)
71.04

Values as printed on the card, with the publisher's own settings (card section: Training, Testing, and Evaluation Datasets: › Evaluation Dataset:); column “Nemotron-3-Embed-1B-BF16”. Not comparable across publishers and not used in any ranking. Model card

Quality × memory

No comparable quality × memory pair

Embedding memory estimates cluster at a few GB, so a quality × memory chart would not separate these models; MTEB is ranked above instead.

Runs on

Fast memory holds it 119

The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.

Show all 119

Hardware profiles grouped by the embedding planning estimate (8-bit weights, 20% allowance, 2 GB reserve). Not a measured fit.

Consider it for

Consider Nemotron 3 Embed 1B for retrieval and semantic search: it turns text into 2,048-dimension vectors.

Check first

The publisher states no memory figure; the hardware list uses a ~4 GB planning estimate, a ceiling and not a measurement. See the prompt format under How to run. PT-BR is confirmed by a source.

Measured speed

No published measurement for this checkpoint

Throughput (texts per second) is shown only when a source publishes it with hardware, build and setup. It is never estimated.

How to run

The planning estimate uses 8-bit weights, 20% working allowance and a 2 GB reserve; it is a ceiling, not a measurement.

Runs in
Sentence Transformers, vLLM, PyTorchsource
Query/document prompts
Add the 'query: ' prefix for queries and the 'passage: ' prefix for documents. Embeddings are L2-normalised.
Languages · PT-BR
Evaluated on 34 languages, Portuguese includedPT-BR confirmed by a source

Documented ways to run it

Copied as printed in the linked source (read 2026-10-18); not run by this site.

Setup stated first by the source · Sentence Transformers
pip install --upgrade "transformers>=5.2.0" "sentence-transformers>=5.4.1"
Sentence Transformers · model card · Sentence Transformers
import torch
from sentence_transformers import SentenceTransformer

MODEL_ID = "nvidia/Nemotron-3-Embed-1B-BF16"

model = SentenceTransformer(
    MODEL_ID,
    device="cuda",
    model_kwargs={
        "dtype": torch.bfloat16,
        "attn_implementation": "flash_attention_2",
    },
)
model.max_seq_length = 32768

QUERIES = [
    "Write a Python function that counts the frequency of each element in a list of lists.",
    "Write a function that orders a dictionary with tuple keys by the product of each key's tuple values.",
    "What symptoms and common triggers help distinguish eczema from other inflammatory skin conditions?",
    "How can someone reduce exposure to pollen during allergy season?",
]

DOCUMENTS = [
    "def frequency_lists(list1):\n    flattened = [item for sublist in list1 for item in sublist]\n    counts = {}\n    for item in flattened:\n        if item in counts:\n            counts[item] += 1\n        else:\n            counts[item] = 1\n    return counts",
    "def sort_dict_item(test_dict):\n    return {key: test_dict[key] for key in sorted(test_dict.keys(), key=lambda ele: ele[0] * ele[1])}",
    "Eczema commonly causes itchy, dry, inflamed patches of skin. The affected areas may look red, scaly, cracked, or darker than the surrounding skin depending on skin tone. Symptoms can flare after exposure to irritants, allergens, stress, or changes in weather.",
    "People with pollen allergy can reduce exposure by staying indoors on dry, windy days, avoiding early-morning outdoor activity, and going outside after rain when pollen levels are lower. They should check pollen forecasts, close windows and doors when counts are high, and consider starting allergy medication before symptoms begin if high pollen is expected. After being outside, showering, changing clothes, avoiding outdoor laundry drying, and wearing a face mask for yard work can help limit pollen contact.",
]
query_embeddings = model.encode_query(QUERIES, batch_size=8, convert_to_tensor=True)
document_embeddings = model.encode_document(DOCUMENTS, batch_size=8, convert_to_tensor=True)

scores = model.similarity(query_embeddings, document_embeddings)
print("Similarity scores:")
print(f"{'':>4}" + "".join(f"d[{i}]".rjust(10) for i in range(scores.shape[1])))
for query_index, row in enumerate(scores):
    print(f"q[{query_index}]" + "".join(f"{score.item():>10.4f}" for score in row))
Setup stated first by the source · vLLM
pip install --upgrade "vllm==0.25.0" openai requests numpy
vLLM · model card · vLLM Online Serving
MODEL_ID=nvidia/Nemotron-3-Embed-1B-BF16

vllm serve "$MODEL_ID"
Setup stated first by the source · Transformers
pip install --upgrade "transformers>=5.2.0" "sentence-transformers>=5.4.1"
Transformers · model card · Transformers
import torch
import torch.nn.functional as F
from transformers import AutoModel, AutoTokenizer

MODEL_ID = "nvidia/Nemotron-3-Embed-1B-BF16"
MAX_LENGTH = 32768
BATCH_SIZE = 8
DTYPE = torch.bfloat16
ATTN_IMPLEMENTATION = "flash_attention_2"

QUERIES = [
    "Write a Python function that counts the frequency of each element in a list of lists.",
    "Write a function that orders a dictionary with tuple keys by the product of each key's tuple values.",
    "What symptoms and common triggers help distinguish eczema from other inflammatory skin conditions?",
    "How can someone reduce exposure to pollen during allergy season?",
]

DOCUMENTS = [
    "def frequency_lists(list1):\n    flattened = [item for sublist in list1 for item in sublist]\n    counts = {}\n    for item in flattened:\n        if item in counts:\n            counts[item] += 1\n        else:\n            counts[item] = 1\n    return counts",
    "def sort_dict_item(test_dict):\n    return {key: test_dict[key] for key in sorted(test_dict.keys(), key=lambda ele: ele[0] * ele[1])}",
    "Eczema commonly causes itchy, dry, inflamed patches of skin. The affected areas may look red, scaly, cracked, or darker than the surrounding skin depending on skin tone. Symptoms can flare after exposure to irritants, allergens, stress, or changes in weather.",
    "People with pollen allergy can reduce exposure by staying indoors on dry, windy days, avoiding early-morning outdoor activity, and going outside after rain when pollen levels are lower. They should check pollen forecasts, close windows and doors when counts are high, and consider starting allergy medication before symptoms begin if high pollen is expected. After being outside, showering, changing clothes, avoiding outdoor laundry drying, and wearing a face mask for yard work can help limit pollen contact.",
]

def average_pool(last_hidden_state: torch.Tensor, attention_mask: torch.Tensor) -> torch.Tensor:
    last_hidden = last_hidden_state.masked_fill(~attention_mask[..., None].bool(), 0.0)
    return last_hidden.sum(dim=1) / attention_mask.sum(dim=1)[..., None]

if not torch.cuda.is_available():
    raise RuntimeError("CUDA is required for practical BF16 inference.")

device = torch.device("cuda")

tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, padding_side="left")
if tokenizer.pad_token is None:
    tokenizer.pad_token = tokenizer.eos_token

model = AutoModel.from_pretrained(
    MODEL_ID,
    dtype=DTYPE,
    attn_implementation=ATTN_IMPLEMENTATION,
).to(device)
model.eval()

def encode_texts(texts: list[str]) -> torch.Tensor:
    embedding_batches = []

    for start in range(0, len(texts), BATCH_SIZE):
        encoded = tokenizer(
            texts[start : start + BATCH_SIZE],
            max_length=MAX_LENGTH,
            truncation=True,
            padding=True,
            return_tensors="pt",
        )
        encoded = {name: tensor.to(device) for name, tensor in encoded.items()}

        with torch.inference_mode():
            output = model(**encoded)
            pooled = average_pool(output.last_hidden_state, encoded["attention_mask"])
            embeddings = F.normalize(pooled, p=2, dim=-1)

        embedding_batches.append(embeddings.detach().cpu().to(torch.float32))

    return torch.cat(embedding_batches, dim=0)

embeddings = encode_texts(
    ["query: " + query for query in QUERIES]
    + ["passage: " + doc for doc in DOCUMENTS]
)
query_embeddings = embeddings[: len(QUERIES)]
document_embeddings = embeddings[len(QUERIES) :]

scores = query_embeddings @ document_embeddings.T
print("Similarity scores:")
print(f"{'':>4}" + "".join(f"d[{i}]".rjust(10) for i in range(scores.shape[1])))
for query_index, row in enumerate(scores):
    print(f"q[{query_index}]" + "".join(f"{score.item():>10.4f}" for score in row))

Choose my computer

Technical details

What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.

Task
embeddings for search (feature extraction)
Library in source
Sentence Transformers, vLLM, PyTorch
Weight license
OpenMDW 1.1
Commercial use
Allowed OpenMDW 1.1 text (nvidia/NV-Reason-CT LICENSE) · read 2026-10-18license text
Parameter basis
Hub safetensors total (1,140,918,272 parameters, read 2026-10-09 from the Hub API).
Announced
2026-07-16

Sources

Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.

Curated facts checked: 2026-10-09. Each value cites its source (the publisher's card or the MTEB results dataset); estimates are labelled as estimates, and a value no source states stays empty. A repository commit is not a model release date.