Embeddings API

What Are Embeddings?

Embeddings are vector representations of pieces of texts. If 2 pieces of text have a similar vector representation, it most likely means that they have a similar meaning.

Imagine you have the 3 following sentences:

NLP Cloud is an API for natural language processing.

NLP Cloud proposes an API dedicated to NLP at scale.

I went to the cinema yesterday. It was great!

Here are the embeddings from the 3 above sentences (truncated for the sake of simplicity):

[[0.0927242711186409,-0.19866740703582764,-0.013638739474117756,-0.11876793205738068,0.011521861888468266,-0.03629707545042038, -0.030676838010549545,-0.03159608319401741,0.021390020847320557,0.03344911336898804,0.1698218137025833,-0.0009996045846492052, -0.07465217262506485,-0.21483412384986877,0.11283198744058609,0.03549865633249283,0.04985387250781059,-0.027558118104934692, 0.06297887861728668,0.09421529620885849,0.03700404614210129,0.06565431505441666,0.02284885197877884,0.06327767670154572, -0.09266531467437744,-0.014569456689059734,-0.06129194051027298,0.1818675994873047,0.09628438949584961,-0.09874546527862549, 0.030865425243973732, [...] ,-0.02097163535654545,0.021617714315652847,0.11045169830322266,0.01000999379903078,0.11451057344675064,0.18813028931617737, 0.007419265806674957,0.1630171686410904,0.21308083832263947,-0.03355317562818527,0.0778832957148552,0.2268853485584259,-0.13271427154541016, 0.005264544393867254,0.16081497073173523,0.09937280416488647,-0.12673905491828918,-0.12035898119211197,-0.06462062895298004, -0.0024213052820414305,0.08730605989694595,-0.04702030122280121,-0.03694896399974823,0.002265638206154108,-0.027780283242464066, -0.00017151003703474998,-0.20887477695941925,-0.2585527300834656,0.3124837279319763,0.05403835326433182,0.027094876393675804, -0.022925367578864098,0.038322173058986664]]

Embeddings are a core feature of Natural Language Processing because, once a machine is able to detect similarities between texts, it paves the ways for many interesting applications like semantic similarity, RAG (retrieval augmented generation) systems, semantic search, paraphrase detection, clustering, and more.

AI Embeddings

Why Extract Embeddings?

Here are some examples where embeddings are extremely useful:

Semantic Similarity

You might want to detect whether 2 sentences are talking about the same thing or not. That's useful for paraphrase (plagiarism) detection for example. It's also useful to understand if several persons are talking about the same topic or not.

Semantic Search

Semantic search is the modern way of searching for information. Instead of naively searching for texts containing specific keywords, you can now search for texts talking about a topic you're interested in, even if keywords don't match (in case of synonyms for examples).


You might want to group things by categories (ideas, speeches, conversations...). Clustering is an old machine learning technique that can now be effectively applied to natural language processing.

RAG Systems

RAG (Retrieval Augmented Generation) systems are a type of natural language processing model that generates text by combining the capabilities of a large-scale language model with a retrieval component that fetches relevant information from a database or corpus of texts. This approach allows the generation of more accurate, informative, and contextually relevant responses by leveraging external knowledge sources.

NLP Cloud's Embeddings API

NLP Cloud proposes an embeddings API that gives you the opportunity to extract embeddings out of the box, based on Sentence Transformers models like Paraphrase Multilingual Mpnet Base v2.
The response time (latency) is very low for embeddings models, which allows you to easily include embeddings extraction into a larger and more complex workflow.

For more details, see our documentation about embeddings here.

Testing embeddings locally is one thing, but using it reliably in production is another thing. With NLP Cloud you can just do both!

Frequently Asked Questions

Why are embeddings important in machine learning and AI?

Embeddings are crucial in machine learning and AI because they enable the representation of high-dimensional, sparse data (like words, images, or user behaviors) in a dense, lower-dimensional space, preserving semantic relationships and patterns. This facilitates more efficient and effective learning by models, allowing for improved performance on tasks such as classification, recommendation, and natural language understanding.

How can one evaluate the quality of embeddings?

The quality of embeddings can be evaluated through intrinsic methods, such as analogy tasks or clustering evaluations that directly measure the embeddings' representation of linguistic or conceptual relationships. Alternatively, extrinsic evaluation methods assess the improvement in performance of downstream tasks, like text classification or sentiment analysis, when using the embeddings.

How are embeddings used in recommendation systems?

In recommendation systems, embeddings are used to convert items and users into vectors in a lower-dimensional space, capturing complex patterns and preferences. By computing similarity measures between these vectors, the system can efficiently recommend items likely to appeal to a user based on their historical interactions and the interactions of others with similar tastes.

What are contextual embeddings and why are they important?

Contextual embeddings are advanced representations of words that capture the meaning based on the surrounding text, unlike static embeddings which assign a single embedding to each word regardless of its context. They are important because they allow models to understand nuances in language, such as homonyms or words that change meaning based on the surrounding words, leading to more accurate interpretations in natural language processing tasks. This is done by default on NLP Cloud.

How are embeddings useful in a RAG system?

In a Retrieval-Augmented Generation (RAG) system, embeddings are crucial for effectively retrieving relevant documents or data entries from a large corpus, based on the semantic similarity to a given query. This retrieval step enriches the input to the generation component, leading to more informed, accurate, and contextually appropriate responses or content generation.

How are embeddings useful in semantic search?

Embeddings are useful in semantic search as they convert text into dense vectors that capture the semantic meaning and relationships between words or phrases, enabling the search algorithm to understand and retrieve content that is contextually relevant to the query, even if the exact keywords are not present. This significantly enhances the accuracy and relevance of search results by focusing on the intent and meaning behind the user's query rather than relying solely on keyword matching.

Can I try the embeddings API for free?

Yes, like all the models on NLP Cloud, the embeddings API endpoint can be tested for free

How does your AI API handle data privacy and security during the embeddings extraction process?

NLP Cloud is focused on data privacy by design: we do not log or store the content of the requests you make on our API. NLP Cloud is both HIPAA and GDPR compliant.

What are the supported languages or locales for this embeddings API?

Our embeddings API supports 50 languages