Lepton AI

What Lepton AI does

Lepton AI was a Palo Alto-based, cloud-native GPU compute rental platform founded in 2023 by Jia Yangqing, former Vice President at Alibaba who led the company's computing platform division. The company provided serverless GPU endpoints that gave AI developers, data scientists, and model-building teams on-demand access to NVIDIA GPU infrastructure for training, inference, and general AI workloads, abstracting away capital expenditure and operational complexity associated with owning and managing GPU clusters. Its core technology centered on a proprietary orchestration layer enabling multi-cloud GPU provisioning and instant, usage-based access to compute across providers and regions.

The business model was usage-based compute rental: developers and AI teams paid for GPU capacity consumed, with no upfront hardware commitment. The go-to-market was API-first and developer-led, targeting a horizontal segment of AI practitioners rather than named enterprise accounts. Pricing was not publicly disclosed. The company operated with a small team (11-50 employees) and completed only an angel funding round in May 2023 with CRV and Fusion Fund participating.

In March 2025, NVIDIA acquired Lepton AI for a reported several hundred million dollars as part of its strategy to diversify beyond GPU hardware into AI cloud services and enterprise software. Following the acquisition, Lepton's platform technology was integrated into NVIDIA DGX Cloud Lepton, extending distribution through NVIDIA's global cloud partner network and the build.nvidia.com marketplace, and combining with NVIDIA NIM microservices and serverless endpoints to form a broader multi-cloud AI compute offering.

Name

Lepton AI

Legal name

Lepton AI

Website

https://lepton.ai

Company type

Private

Founded year

2023

Operating status

Acquired

Headcount range

1–10 employees

Ownership category

akta.pro rank

Product category

Cloud GPU Infrastructure

NAICS

Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182), Computer Systems Design and Related Services (54151)

SIC

Services-Computer Rental & Leasing (7377), Services-Computer Integrated Systems Design (7373)

akta.pro primary industry

AI Compute Cloud & GPU-as-a-Service (HDAAAAAK)

akta.pro secondary industries

AI Compute Virtualization & Scheduling (GPU virtualization, cluster schedulers) (HDAAAAAG), AI Server Systems & HGX/Accelerator Platforms (HDAAAAAB), GPU-Accelerated & AI Training/Inference Servers (HDACABAG)

Keywords

• GPU cloud rental
• Serverless GPU endpoints
• AI compute platform
• Cloud GPU access
• AI infrastructure rental

Headquarters

HQ city

Palo Alto

HQ country

United States

HQ region

North America

Markets served

GTM type

B2B

Offering type

Software

Cost components

Infrastructure, Technology or R&D, Personnel, Marketing or Sales

Revenue model

1. GPU Compute Rental: Lepton AI rented out NVIDIA GPU servers to AI developers and companies, providing on-demand access to GPU compute capacity. Revenue was generated through usage-based compute rental fees.

Go-to-market motion1 record

Distribution channels1 record

Core offering

Lepton AI operated a cloud-native platform for renting NVIDIA GPU servers to AI developers and companies, providing serverless GPU endpoints and on-demand compute access for AI model training, development, and inference workloads. Following its 2025 acquisition by NVIDIA, the technology was integrated into NVIDIA DGX Cloud Lepton, offering unified multi-cloud GPU access with instant endpoints and prebuilt NVIDIA NIM microservices.

Differentiator

Problem solved

Functional benefit

Products and services

• Lepton AI GPU Cloud Rental Platform A pre-acquisition GPU cloud rental platform that rented out NVIDIA GPU servers to AI developers and companies, providing on-demand serverless GPU endpoints for AI model training, development, and inference. Targeted at AI developers and model builders who needed elastic GPU compute without managing infrastructure.
• NVIDIA DGX Cloud Lepton The post-acquisition integration of Lepton AI's GPU rental platform into NVIDIA's cloud services. Connects developers to global GPU compute across multiple cloud providers and regions, providing instant access to accelerated APIs including serverless endpoints and prebuilt NVIDIA NIM microservices through build.nvidia.com.

Named customers1 record

Segments1 record

Ideal customer profiles1 record

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

Feature2 records

Scale indicators2 records

Recent moves5 records

Expansion highlights5 records

Direct peers

• CoreWeave: CoreWeave is a large-scale GPU cloud provider offering on-demand NVIDIA GPU compute for AI training and inference. It is a direct competitor to Lepton AI's pre-acquisition GPU rental platform and one of the leading independent players in the same category.
• Lambda Labs: Lambda Labs provides GPU cloud instances and clusters purpose-built for deep learning, directly competing with Lepton AI's GPU rental and serverless endpoints serving AI developers and researchers.
• Together AI: Together AI operates a cloud platform for open-source and frontier AI models with GPU-accelerated inference and training, competing directly with Lepton's serverless GPU endpoints targeting AI developers.
• Modal Labs: Modal provides a serverless cloud platform for running AI and data workloads on GPUs with a developer-first API, very similar to Lepton AI's serverless GPU endpoint offering.
• Replicate: Replicate runs machine learning models in the cloud via a simple API on GPU-backed infrastructure, overlapping directly with Lepton AI's developer-facing serverless GPU compute value proposition.
• RunPod: RunPod offers on-demand GPU cloud instances and serverless GPU endpoints for AI workloads, closely matching Lepton AI's pre-acquisition product offering of serverless GPU access for AI developers.

Broad incumbents

• Amazon Web Services (EC2 P-series / SageMaker): AWS offers GPU-accelerated EC2 instances, SageMaker, and Bedrock for AI workloads as part of its broad cloud portfolio. It competes with Lepton AI for AI developer compute but as a hyperscaler incumbent rather than a focused GPU cloud.
• Google Cloud Platform (GCP GPU / Vertex AI): Google Cloud provides GPU instances, Vertex AI, and serverless inference offerings that overlap with Lepton AI's GPU rental and serverless AI endpoints within a much broader cloud platform.
• Microsoft Azure (Azure ML / GPU VMs): Microsoft Azure offers GPU virtual machines, Azure Machine Learning, and serverless inference endpoints that compete with Lepton AI's developer-facing GPU compute services as part of a large enterprise cloud portfolio.

Emerging players

• Vast.ai: Vast.ai operates a decentralized GPU marketplace where developers can rent GPU compute from third-party hosts, offering a lower-cost alternative to Lepton AI's centralized GPU rental platform with similar target customers.

Market position

Strengths4 records

Weaknesses4 records

Competitive moat3 records

Key risks5 records

Key highlights5 records

Customer concentration

• Xx.com/leptonai
• LinkedInlinkedin.com/company/lepton-ai

Revenue estimate

Valuation estimate

Number of profiles

Profiles1 record

Funding overview

Funding rounds1 record

Investors2 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Lepton AI