nomic-embed-vision-v1.5: Expanding the Latent Space#
Quick Start#
Blog | Technical Report | AWS SageMaker | Atlas Embedding and Unstructured Data Analytics Platform
nomic-embed-vision-v1.5 is a high performing vision embedding model that shares the same embedding space as nomic-embed-text-v1.5 .
All Nomic Embed Text models are now multimodal !
Name Imagenet 0-shot Datacomp (Avg. 38) MTEB nomic-embed-vision-v1.571.0 56.8 62.28 nomic-embed-vision-v170.7 56.7 62.39 OpenAI CLIP ViT B/16 68.3 56.3 43.82 Jina CLIP v1 59.1 52.2 60.1
Hosted Inference API#
The easiest way to get started with Nomic Embed is through the Nomic Embedding API.
Generating embeddings with the nomic Python client is as easy as
from nomic import embed
import numpy as np
output = embed.image(
images=[
"image_path_1.jpeg",
"image_path_2.png",
],
model='nomic-embed-vision-v1.5',
)
print(output['usage'])
embeddings = np.array(output['embeddings'])
print(embeddings.shape)For more information, see the API reference
Data Visualization#
Click the Nomic Atlas map below to visualize a 100,000 sample CC3M comparing the Vision and Text Embedding Space!
image/webp
Training Details#
We align our vision embedder to the text embedding by employing a technique similar to LiT but instead lock the text embedder!
For more details, see the Nomic Embed Vision Technical Report (soon to be released!) and corresponding blog post
Training code is released in the contrastors repository
Usage#
Remember nomic-embed-text requires prefixes and so, when using Nomic Embed in multimodal RAG scenarios (e.g. text to image retrieval), you should use the search_query: prefix.
import torch
import torch.nn.functional as F
from transformers import AutoTokenizer, AutoModel, AutoImageProcessor
from PIL import Image
import requests
processor = AutoImageProcessor.from_pretrained("nomic-ai/nomic-embed-vision-v1.5")
vision_model = AutoModel.from_pretrained("nomic-ai/nomic-embed-vision-v1.5", trust_remote_code=True)
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
inputs = processor(image, return_tensors="pt")
img_emb = vision_model(**inputs).last_hidden_state
img_embeddings = F.normalize(img_emb[:, 0], p=2, dim=1)Additionally, you can perform multimodal retrieval!
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 are cute animals to cuddle with?', 'search_query: What do cats look like?']
tokenizer = AutoTokenizer.from_pretrained('nomic-ai/nomic-embed-text-v1.5')
text_model = AutoModel.from_pretrained('nomic-ai/nomic-embed-text-v1.5', trust_remote_code=True)
text_model.eval()
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
with torch.no_grad():
model_output = text_model(**encoded_input)
text_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
text_embeddings = F.layer_norm(text_embeddings, normalized_shape=(text_embeddings.shape[1],))
text_embeddings = F.normalize(text_embeddings, p=2, dim=1)
print(torch.matmul(img_embeddings, text_embeddings.T))
unsloth 
mlx-community