Andromeda
Andromeda operates a two-sided GPU compute marketplace connecting AI startups and enterprises with 100+ infrastructure providers for on-demand, standardized access to high-performance compute as an alternative to multi-year hyperscaler contracts.
- Company typePrivate
- Founded2022
- HeadquartersSan Francisco, United States
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
What Andromeda does
Andromeda (legally Andromeda Cluster, Inc.) is a San Francisco-headquartered GPU compute marketplace founded in 2022 by Nat Friedman and Daniel Gross, originating from their venture firm NFDG. The company operates a two-sided platform connecting AI teams, research labs, and enterprises seeking high-performance compute with infrastructure providers offering GPU capacity, positioning itself as an on-demand alternative to multi-year hyperscaler contracts. Its customer base spans early-stage AI startups through larger enterprises needing flexible access to cutting-edge GPUs.
The platform consists of a Customer Portal—where buyers specify GPU type, quantity, region, and timeline—and a Provider Portal—where infrastructure owners onboard clusters via Kubernetes or Slurm interfaces and undergo automated certification. Andromeda's core technology includes a real-time GPU benchmarking engine that prices compute across 100+ providers, an automated provider certification engine, and a standardized SLA and contract framework that consolidates procurement into one invoice and one support channel. The marketplace supports H100, H200, B200, B300, GB200, and GB300 GPUs, has facilitated 1,000+ transactions representing billions of GPU/hour in liquidity, and currently manages tens of thousands of GPUs with stated plans to scale to 100,000.
Andromeda generates revenue via transaction fees and usage-based GPU/hour billing on compute provisioned through the marketplace. Go-to-market is primarily product-led through self-serve onboarding for both buyers and providers, supplemented by a sales-led motion targeting larger AI companies. The company holds SOC 2 Type II and ISO 27001 certifications. Andromeda reached a $100 million revenue run rate in 2025, up from over $50 million the prior year, while operating profitably since inception; in March 2026 it secured funding from Paradigm at a $1.5 billion valuation, achieving unicorn status.
Andromeda firmographics
Firmographics- Name
- Andromeda
- Legal name
- Andromeda Cluster, Inc.
- Website
- https://andromeda.ai
- Company type
- Private
- Founded year
- 2022
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- Andromeda operates a two-sided GPU compute marketplace connecting AI startups and enterprises with 100+ infrastructure providers for on-demand, standardized access to high-performance compute as an alternative to multi-year hyperscaler contracts.
- Ownership category
- akta.pro rank
Andromeda industry classification
Industry- Product category
- GPU Compute Marketplace
- NAICS
- Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (518), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (518210)
- SIC
- 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), GPU/Accelerator Compute Servers (HDABADAC)
Keywords
Where Andromeda is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Markets served
Andromeda business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations
Revenue model
- GPU Compute Marketplace Transaction Fees: Platform facilitates transactions between GPU buyers (AI teams seeking compute) and sellers (providers with GPU capacity). The company earns revenue from each transaction completed on the platform, functioning as a marketplace intermediary for high-performance compute resources.
- GPU Capacity Provision: Manages and provisions tens of thousands of GPUs for AI startups and labs, positioning itself as infrastructure for a future GPU commodity market. Revenue generated from compute usage fees.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Usage-based | Pay-as-you-go | GPU compute access priced per GPU/hour with no multi-year commitment required |
Go-to-market motion2 records
Distribution channels3 records
Marketing channels4 records
Andromeda product offering
Product offeringCore offering
Andromeda operates a global GPU compute marketplace platform that connects AI teams and enterprises seeking high-performance compute resources with infrastructure providers offering unused GPU capacity. The platform sources, benchmarks, and prices compute across 100+ providers in real time, offering standardized SLAs and contracts so customers can procure GPU compute without multi-year hyperscaler commitments.
Product overview
Andromeda operates a unified GPU compute marketplace platform consisting of a Customer Portal for AI teams to purchase compute and a Provider Portal for infrastructure owners to sell GPU capacity. The platform sources, benchmarks, and prices compute across 100+ providers in real time, offering GPU access as an alternative to multi-year hyperscaler contracts. The company manages tens of thousands of GPUs and plans to scale to 100,000 GPUs, positioning itself as market infrastructure for a future GPU commodity market.
Differentiator
Problem solved
Functional benefit
Products and services
- GPU Compute Marketplace Platform A marketplace platform connecting AI teams seeking high-performance compute resources with infrastructure providers offering GPU capacity. It sources, benchmarks, and prices compute across 100+ providers in real time and enables transactions on standardized SLAs and contracts. Customers include AI startups, labs, and enterprises; the counterparty side is GPU infrastructure providers.
- Customer Onboarding Portal A self-service intake portal where AI teams and enterprises specify GPU type, quantity, desired start date, and duration to request compute on standardized SLAs and contracts.
- Provider Onboarding Portal A self-service portal for GPU infrastructure providers to register clusters, submit GPU specs (H100/H200/B200/B300/GB200/GB300), pricing, networking details, and availability, and receive qualified demand routed from the marketplace.
Quantifiable outcome
- Revenue run rate of $100 million in 2025, up from over $50 million in 2024
- +2 more outcomes
Companies that use Andromeda
Customer profileSegments3 records
Ideal customer profiles2 records
Andromeda technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- No
- API docs
- API detail
Core technology
AI maturity
App detail
Feature4 records
Andromeda partnerships and signals
Strategic signalScale indicators8 records
Recent moves4 records
Expansion highlights4 records
Andromeda competitors and assessment
Company assessmentDirect peers
- RunPod: RunPod offers on-demand GPU cloud and serverless AI inference targeting AI developers and startups, competing with Andromeda on flexible, no-commitment GPU access.
- Vast.ai: Vast.ai operates a peer-to-peer GPU marketplace aggregating consumer and prosumer GPU capacity, sharing Andromeda's two-sided marketplace model and AI-customer focus.
- CoreWeave: CoreWeave is a specialized GPU cloud provider offering NVIDIA H100/B200/GB200 capacity to AI labs and enterprises, directly competing with Andromeda's GPU compute marketplace by aggregating third-party capacity and reselling at scale.
- Lambda Labs: Lambda operates GPU cloud infrastructure and on-demand clusters for AI training and inference, targeting the same AI startup and lab customer base as Andromeda with usage-based GPU pricing.
- Crusoe Energy: Crusoe provides large-scale GPU cloud capacity for AI workloads, often sourced from energy-flared or stranded power sites, competing directly with Andromeda in the AI compute provisioning space.
Broad incumbents
- Google Cloud (GPU / A3 instances): Google Cloud provides H100/A3 GPU instances integrated with its broader data, AI, and analytics stack, competing with Andromeda for enterprise AI compute workloads.
- Amazon Web Services (EC2 GPU instances): AWS offers P4/P5 GPU instance types for AI training and inference as part of a broader cloud portfolio, representing the dominant incumbent that Andromeda's flexible, no-commitment marketplace positions against.
- Microsoft Azure (ND H100 v5 series): Azure offers ND-series GPU VMs for AI workloads with tight integration into the Microsoft enterprise and OpenAI ecosystem, competing with Andromeda for hyperscale AI training customers.
Emerging players
- Together AI: Together AI provides a managed AI cloud platform combining GPU compute with model inference and fine-tuning services, overlapping with Andromeda on GPU provisioning for AI startups while adding a higher-level application stack.
- Fluidstack: Fluidstack is an AI cloud provider offering large-scale GPU clusters to AI labs, often partnering with energy providers, overlapping with Andromeda's marketplace model of aggregating third-party GPU capacity.
Market position
Strengths4 records
Weaknesses4 records
Competitive moat4 records
Key risks5 records
Key highlights6 records
Customer concentration
Andromeda social profiles
Digital presenceAndromeda compliance and trust
Trust signalCompliance2 records
Andromeda financial estimates
Financial estimateRevenue estimate
Valuation estimate
Andromeda leadership team
Management profileNumber of profiles
Profiles2 records
Andromeda funding detail
Funding detailFunding overview
Funding rounds2 records
Investors1 record
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Andromeda M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Andromeda
What does Andromeda do?
Andromeda operates a global GPU compute marketplace platform that connects AI teams and enterprises seeking high-performance compute resources with infrastructure providers offering unused GPU capacity. The platform sources, benchmarks, and prices compute across 100+ providers in real time, offering standardized SLAs and contracts so customers can procure GPU compute without multi-year hyperscaler commitments.
Is Andromeda a public or private company?
Andromeda is a private company. It is classified as venture growth investor backed and is currently operating.
When was Andromeda founded?
Andromeda was founded in 2022. It employs 11 to 50 people.
Where is Andromeda based?
Andromeda is headquartered in San Francisco, United States, in the North America region.
How does Andromeda make money?
Two revenue lines are on record. GPU Compute Marketplace Transaction Fees are the primary driver. The others are GPU Capacity Provision.
Who are Andromeda's main competitors?
Direct peers on record are RunPod, Vast.ai, CoreWeave, Lambda Labs and Crusoe Energy. Broad incumbents are Google Cloud (GPU / A3 instances), Amazon Web Services (EC2 GPU instances) and Microsoft Azure (ND H100 v5 series). Emerging players are Together AI and Fluidstack.
Does Andromeda have an API?
No public API is recorded for Andromeda.
What industry is Andromeda in?
Andromeda's product category is GPU Compute Marketplace. Its primary akta.pro industry code is HDAAAAAK, AI Compute Cloud & GPU-as-a-Service, with a secondary code of HDAAAAAG, AI Compute Virtualization & Scheduling (GPU virtualization, cluster schedulers). Its NAICS code is 518 and its SIC code is 7373.