nomic-embed-text-v1.5: Resizable Production Embeddings with Matryoshka Representation Learning

Exciting Update!: nomic-embed-text-v1.5 is now multimodal! nomic-embed-vision-v1 is aligned to the embedding space of nomic-embed-text-v1.5, meaning any text embedding is multimodal!

Usage

Important: the text prompt must include a task instruction prefix, instructing the model which task is being performed.

For example, if you are implementing a RAG application, you embed your documents as search_document: <text here> and embed your user queries as search_query: <text here>.

Task instruction prefixes

search_document

Purpose: embed texts as documents from a dataset

This prefix is used for embedding texts as documents, for example as documents for a RAG index.

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("nomic-ai/nomic-embed-text-v1", trust_remote_code=True)
sentences = ['search_document: TSNE is a dimensionality reduction algorithm created by Laurens van Der Maaten']
embeddings = model.encode(sentences)
print(embeddings)

search_query

Purpose: embed texts as questions to answer

This prefix is used for embedding texts as questions that documents from a dataset could resolve, for example as queries to be answered by a RAG application.

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("nomic-ai/nomic-embed-text-v1", trust_remote_code=True)
sentences = ['search_query: Who is Laurens van Der Maaten?']
embeddings = model.encode(sentences)
print(embeddings)

clustering

Purpose: embed texts to group them into clusters

This prefix is used for embedding texts in order to group them into clusters, discover common topics, or remove semantic duplicates.

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("nomic-ai/nomic-embed-text-v1", trust_remote_code=True)
sentences = ['clustering: the quick brown fox']
embeddings = model.encode(sentences)
print(embeddings)

classification

Purpose: embed texts to classify them

This prefix is used for embedding texts into vectors that will be used as features for a classification model

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("nomic-ai/nomic-embed-text-v1", trust_remote_code=True)
sentences = ['classification: the quick brown fox']
embeddings = model.encode(sentences)
print(embeddings)

Sentence Transformers

import torch.nn.functional as F
from sentence_transformers import SentenceTransformer

matryoshka_dim = 512

model = SentenceTransformer("nomic-ai/nomic-embed-text-v1.5", trust_remote_code=True)
sentences = ['search_query: What is TSNE?', 'search_query: Who is Laurens van der Maaten?']
embeddings = model.encode(sentences, convert_to_tensor=True)
embeddings = F.layer_norm(embeddings, normalized_shape=(embeddings.shape[1],))
embeddings = embeddings[:, :matryoshka_dim]
embeddings = F.normalize(embeddings, p=2, dim=1)
print(embeddings)

Transformers

import torch
import torch.nn.functional as F
from transformers import AutoTokenizer, AutoModel

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)

sentences = ['search_query: What is TSNE?', 'search_query: Who is Laurens van der Maaten?']

tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
model = AutoModel.from_pretrained('nomic-ai/nomic-embed-text-v1.5', trust_remote_code=True, safe_serialization=True)
model.eval()

encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

+ matryoshka_dim = 512

with torch.no_grad():
    model_output = model(**encoded_input)

embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
+ embeddings = F.layer_norm(embeddings, normalized_shape=(embeddings.shape[1],))
+ embeddings = embeddings[:, :matryoshka_dim]
embeddings = F.normalize(embeddings, p=2, dim=1)
print(embeddings)

The model natively supports scaling of the sequence length past 2048 tokens. To do so,

- tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
+ tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased', model_max_length=8192)


- model = AutoModel.from_pretrained('nomic-ai/nomic-embed-text-v1', trust_remote_code=True)
+ model = AutoModel.from_pretrained('nomic-ai/nomic-embed-text-v1', trust_remote_code=True, rotary_scaling_factor=2)

Transformers.js

import { pipeline, layer_norm } from '@xenova/transformers';

// Create a feature extraction pipeline
const extractor = await pipeline('feature-extraction', 'nomic-ai/nomic-embed-text-v1.5', {
    quantized: false, // Comment out this line to use the quantized version
});

// Define sentences
const texts = ['search_query: What is TSNE?', 'search_query: Who is Laurens van der Maaten?'];

// Compute sentence embeddings
let embeddings = await extractor(texts, { pooling: 'mean' });
console.log(embeddings); // Tensor of shape [2, 768]

const matryoshka_dim = 512;
embeddings = layer_norm(embeddings, [embeddings.dims[1]])
    .slice(null, [0, matryoshka_dim])
    .normalize(2, -1);
console.log(embeddings.tolist());

Nomic API

The easiest way to use Nomic Embed is through the Nomic Embedding API.

Generating embeddings with the nomic Python client is as easy as

from nomic import embed

output = embed.text(
    texts=['Nomic Embedding API', '#keepAIOpen'],
    model='nomic-embed-text-v1.5',
    task_type='search_document',
    dimensionality=256,
)

print(output)

For more information, see the API reference

Infinity

Usage with Infinity.

docker run --gpus all -v $PWD/data:/app/.cache -e HF_TOKEN=$HF_TOKEN -p "7997":"7997" \
michaelf34/infinity:0.0.70 \
v2 --model-id nomic-ai/nomic-embed-text-v1.5 --revision "main" --dtype float16 --batch-size 8 --engine torch --port 7997 --no-bettertransformer

Adjusting Dimensionality

nomic-embed-text-v1.5 is an improvement upon Nomic Embed that utilizes Matryoshka Representation Learning which gives developers the flexibility to trade off the embedding size for a negligible reduction in performance.

Name SeqLen Dimension MTEB
nomic-embed-text-v1 8192 768 62.39
nomic-embed-text-v1.5 8192 768 62.28
nomic-embed-text-v1.5 8192 512 61.96
nomic-embed-text-v1.5 8192 256 61.04
nomic-embed-text-v1.5 8192 128 59.34
nomic-embed-text-v1.5 8192 64 56.10

image/png

Training

Click the Nomic Atlas map below to visualize a 5M sample of our contrastive pretraining data!

image/webp

We train our embedder using a multi-stage training pipeline. Starting from a long-context BERT model, the first unsupervised contrastive stage trains on a dataset generated from weakly related text pairs, such as question-answer pairs from forums like StackExchange and Quora, title-body pairs from Amazon reviews, and summarizations from news articles.

In the second finetuning stage, higher quality labeled datasets such as search queries and answers from web searches are leveraged. Data curation and hard-example mining is crucial in this stage.

For more details, see the Nomic Embed Technical Report and corresponding blog post.

Training data to train the models is released in its entirety. For more details, see the contrastors repository

Join the Nomic Community

Citation

If you find the model, dataset, or training code useful, please cite our work

@misc{nussbaum2024nomic,
      title={Nomic Embed: Training a Reproducible Long Context Text Embedder}, 
      author={Zach Nussbaum and John X. Morris and Brandon Duderstadt and Andriy Mulyar},
      year={2024},
      eprint={2402.01613},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
Downloads last month
1,589,252
Safetensors
Model size
137M params
Tensor type
F32
·
Inference Examples
Inference API (serverless) does not yet support model repos that contain custom code.

Model tree for nomic-ai/nomic-embed-text-v1.5

Finetunes
15 models
Quantizations
7 models

Spaces using nomic-ai/nomic-embed-text-v1.5 49

Collection including nomic-ai/nomic-embed-text-v1.5