Nomic Embed Text v2 (MoE)
Nomic AI · nomic-ai/nomic-embed-text-v2-moe
Weights license: Apache 2.0 · code and components may use other terms
Quality
MTEB (Multilingual, v2)
Mean(task) · 0–100 · higher is better · this model outlined
- Color = publisher
Source: MTEB results dataset (independent runs, CC0) · publisher-reported where the model card is the only source · other benchmarks and versions are not mixed in
| MTEB (Multilingual, v2) | mean(task) · 768d | 57.64 | independent run |
|---|---|---|---|
| MTEB (eng, v2) | mean(task) · 768d | 63.64 | independent run |
Reported by the publisher
Values as printed on the card, with the publisher's own settings (card section: Model Overview); column “Nomic Embed v2”. 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.
Consider Nomic Embed Text v2 (MoE) for retrieval and semantic search: it turns text into 768-dimension vectors.
The publisher states no memory figure; the hardware list uses a ~3 GB planning estimate, a ceiling and not a measurement. See the prompt format under How to run. PT-BR is confirmed by a source.
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.
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
- Ollama, Sentence Transformerssource
- Query/document prompts
- Required: “search_query: ” before queries and “search_document: ” before documents (model card).
- Languages · PT-BR
- About 100 languages (model card); Portuguese is in the card's language listPT-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.
pip install torch transformers einops git+https://github.com/nomic-ai/megablocks.gitfrom sentence_transformers import SentenceTransformer
model = SentenceTransformer("nomic-ai/nomic-embed-text-v2-moe", trust_remote_code=True)
sentences = ["Hello!", "¡Hola!"]
embeddings = model.encode(sentences, prompt_name="passage")
print(embeddings.shape)
# (2, 768)
similarity = model.similarity(embeddings[0], embeddings[1])
print(similarity)
# tensor([[0.9118]])pip install torch transformers einops git+https://github.com/nomic-ai/megablocks.gitimport torch
import torch.nn.functional as F
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("nomic-ai/nomic-embed-text-v2-moe")
model = AutoModel.from_pretrained("nomic-ai/nomic-embed-text-v2-moe", trust_remote_code=True)
sentences = ['search_document: Hello!', 'search_document: ¡Hola!']
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0]
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
model.eval()
with torch.no_grad():
model_output = model(**encoded_input)
embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
embeddings = F.normalize(embeddings, p=2, dim=1)
print(embeddings.shape)
# torch.Size([2, 768])
similarity = F.cosine_similarity(embeddings[0], embeddings[1], dim=0)
print(similarity)
# tensor(0.9118)Technical details
What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.
- Checkpoint / source
- nomic-ai/nomic-embed-text-v2-moe
- Task
- embeddings for search (feature extraction)
- Library in source
- Ollama, Sentence Transformers
- Weight license
- Apache 2.0
- Commercial use
- Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
- Parameter basis
- 475M total; the card states 305M active during inference (mixture of experts). Hub safetensors total 475,292,928, read 2026-10-07.
- Released
- Not published by the publisher. Card and the contrastors GitHub README (https://raw.githubusercontent.com/nomic-ai/contrastors/main/README.md) give no release date; the paper date is not a model release date.Checked 2026-10-13.source checked
Ollama's separate “nomic-embed-text” page is v1.5, a different model, and is not used here. The card says Matryoshka dimensions run “from 768 to 256”; only the two endpoints are stored.
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.
- Publisher model card
- Hugging Face model page / card · 2026-10-07
- Hugging Face API (parameters, licence, library) · 2026-10-07
- Ollama library (availability only) · 2026-10-07
- MTEB results dataset (CC0), scores computed by scripts/ingest-mteb-results.mjs · 2026-10-25
Curated facts checked: 2026-10-07. 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.






























































