Transformers,
10x Faster

Ship new NLP features faster on top of over 5,000 pre-trained state of the art models, with 2x to 10x faster inference than out of the box deployment, and scalability built-in.

20 Jay Street, Suite 620
Brooklyn, New York 11201

TourFeaturesPricingSTART FREE TRIAL

Hugging Face is trusted in production by over 5,000 companies

There's a model for that

Leverage the largest repository of fine-tuned Transformer models to solve your business problems with the right tasks, languages and performance.

215 Languages

Understand your customers with thousands of models fine-tuned to 215 languages, or translate text with multi-language models.

NEW YORK

Β© 2021 Hugging Face Inc.. All rights reserved

FOLLOW US

ABOUT US

Hugging Face is a team of scientists and engineers dedicated to democratizing NLP.

Leverage 5,000+ Transformer models

Build your NLP features on top of the largest and fastest-growing collection of State-of-the-Art models.

Reduce user latency serving models in production, with inference up to 10x faster than out-of-box.

Get up to 10x inference speedup

Scale to 1,000 requests per sec

Focus on building new features, not on managing GPUs or scaling infrastructure.

πŸ€—

πŸš€

πŸ“ˆ

Build upon the NLP open source leader

Build your business on a framework-agnostic platform powered by the fastest-growing open source project in NLP.

Do more with NLP

Watch CEO ClΓ©ment Delangue walk through the latest models, and how you can use them to build new features for your customers.

WATCH DEMO

Text Classification

Tag content with Sentiment Analysis, Offensive Language Identification, even with user-created labels using Zero-Shot Classification.

Text Generation

Power up your chatbot experience, summarize documents, or create content automatically with state of the art Transformer models.

Plans & Pricing

Affordable plans to integrate State of the Art NLP models in production, with accelerated inference and scalability built-in.

STARTUP

$599/mo

Everything in LAB 
+
50M input characters
GPU model acceleration
10x inference speedup

5 pinned models

START FREE TRIAL

πŸ‘¨β€πŸ’Ό

ENTERPRISE

Custom

Everything in STARTUP 
+
Dedicated
 endpoints, or
On-premise deployment
Model fine-tuning
Model distillation / pruning
Premium support

CONTACT US

🏒

LAB

5,000+ public models
Private models inference

10M input characters
CPU model acceleration
2x inference speedup

1 pinned model
Email support

START FREE TRIAL

$199/mo

πŸ‘©β€πŸ”¬

PARIS

50 rue Saint George
75009 Paris, France

πŸ‡ΊπŸ‡³

πŸ“„

✍️

❓

Question-Answering

Answer your customer questions, even with a spreadsheet as context! Create your own search knowledge base with sentence embeddings.