{"id":"4631991b-ab40-4993-ab2f-55b75714806f","name":"DigitalOcean","slug":"digitalocean","description":"DigitalOcean is a publicly traded (NYSE: DOCN) developer and small-to-mid-market cloud that entered the GPU market in 2023 by acquiring Paperspace and folding its infrastructure into the DigitalOcean brand as \"GPU Droplets.\" Unlike the training-focused neocloud pack (CoreWeave, Crusoe, Nebius) chasing frontier-lab contracts on 100,000-GPU campuses, DigitalOcean sells GPU capacity the same way it sells VMs: per-second billing, self-serve via API, and single-GPU minimums aimed at startups, indie developers, and enterprises building inference or fine-tuning workloads rather than pre-training runs.\n\nIts GPU fleet is spread across colocation footprint (NYC2, TOR1, ATL1, RIC1, AMS3) rather than owned campuses; a March 2025 expansion deal with Flexential added high-density H200 and AMD Instinct capacity. The catalog covers Nvidia H100, H200, and Blackwell HGX B300 alongside AMD MI300X, MI325X, and MI350X, with on-demand pricing published between roughly $0.76 and $11.19 per GPU-hour. In April 2026 the company repositioned as an \"AI-native cloud built for the inference era\" and launched a managed Inference Engine, tying its GPU business to agentic and inference workloads rather than training scale.\n\nThe strategic bet is the long tail: whereas neoclouds compete on multi-hundred-megawatt anchor tenants, DigitalOcean's advantage is the same self-serve motion, catalog PCI, and roughly 600,000 paying customers that already power its core cloud, with GPU capacity offered as another SKU. Total company revenue was $692.9 million in 2023 (per Wikipedia summary of 10-K); AI-specific revenue is disclosed only qualitatively as a fast-growing segment inside overall results.","status":"active","founded":2011,"hq":"Broomfield, Colorado, USA (incorporated Delaware; historically NYC)","hqLat":null,"hqLng":null,"hqGeocodedAt":null,"hqAddress":null,"hqPrecision":null,"website":"https://www.digitalocean.com","fundingTotal":null,"type":null,"roles":[],"reviewStatus":"reviewed","lifecycleState":"active","supersededByCompanyId":null,"sources":[{"url":"https://www.digitalocean.com/products/gpu-droplets","title":"GPU Droplets product page","publisher":"DigitalOcean"},{"url":"https://www.digitalocean.com/press","date":"2026-05-27","title":"DigitalOcean Newsroom (Hippocratic AI, Inference Engine launch)","publisher":"DigitalOcean"},{"url":"https://en.wikipedia.org/wiki/DigitalOcean","title":"DigitalOcean (company overview, IPO, Paperspace, Flexential partnership)","publisher":"Wikipedia"},{"url":"https://investors.digitalocean.com/","title":"DigitalOcean Investor Relations","publisher":"DigitalOcean"}],"keyFacts":[{"label":"Public status","value":"Public since March 24, 2021 on NYSE under ticker DOCN; S&P 400 component","sourceUrl":"https://en.wikipedia.org/wiki/DigitalOcean"},{"label":"GPU catalog (Nvidia)","value":"HGX B300 (Blackwell), HGX H200, HGX H100, L40S, RTX 6000 Ada, RTX 4000 Ada","sourceUrl":"https://www.digitalocean.com/products/gpu-droplets"},{"label":"GPU catalog (AMD)","value":"Instinct MI350X (CDNA 4), MI325X, MI300X (both CDNA 3); MI350X and B300 gated to sales contact","sourceUrl":"https://www.digitalocean.com/products/gpu-droplets"},{"label":"Announced hourly pricing","value":"On-demand roughly $0.76 to $11.19 per GPU-hour depending on part; reserved commitments quoted $1.91 to $7.94; per-second billing with 5-minute minimum","sourceUrl":"https://www.digitalocean.com/products/gpu-droplets"},{"label":"Datacenter footprint","value":"GPUs available in NYC2, TOR1, ATL1, RIC1, and AMS3 regions; capacity delivered via colocation partners rather than owned campuses","sourceUrl":"https://www.digitalocean.com/products/gpu-droplets"},{"label":"Colocation partner","value":"Flexential (partnership expanded March 2025 for high-density GPU deployments with H200 and AMD Instinct)","sourceUrl":"https://en.wikipedia.org/wiki/DigitalOcean"},{"label":"Paperspace acquisition","value":"Acquired 2023 as the on-ramp to GPU cloud; brand later merged into DigitalOcean GPU Droplets","sourceUrl":"https://en.wikipedia.org/wiki/DigitalOcean"},{"label":"Positioning axis","value":"Inference and fine-tuning oriented (rebranded April 2026 as AI-native cloud for the inference era; launched managed Inference Engine); long-tail SMB rather than hyperscale training","sourceUrl":"https://www.digitalocean.com/press"},{"label":"Named AI customer","value":"Hippocratic AI, disclosed May 2026 as scaling to 10 million patient calls on DigitalOcean using Nvidia Blackwell Ultra GPUs","sourceUrl":"https://www.digitalocean.com/press"},{"label":"Silicon partnerships","value":"Nvidia (H100, H200, B300 Blackwell Ultra) and AMD (MI300X, MI325X, MI350X); no disclosed custom silicon","sourceUrl":"https://www.digitalocean.com/products/gpu-droplets"},{"label":"Total revenue (company-wide)","value":"$692.9M FY2023 (per Wikipedia summary of 10-K); AI/GPU revenue not separately disclosed","sourceUrl":"https://en.wikipedia.org/wiki/DigitalOcean"},{"label":"Employee count","value":"1,204 as of 2022 (Wikipedia summary; company has not published a current figure in the material reviewed)","sourceUrl":"https://en.wikipedia.org/wiki/DigitalOcean"},{"label":"GPU inventory disclosure","value":"Not disclosed; company has not published unit counts of H100/H200/B300/MI300X in reviewed materials","sourceUrl":"https://www.digitalocean.com/products/gpu-droplets"},{"label":"Capacity (MW)","value":"Not disclosed; DigitalOcean does not operate its own datacenters and leases capacity from colocation partners including Flexential","sourceUrl":"https://en.wikipedia.org/wiki/DigitalOcean"}],"aliases":[],"collisionRisk":"low","reviewNote":null,"atsProvider":null,"atsSlug":null,"factoryLocations":null,"manufacturingCapacity":null,"orgType":null,"burnSignal":null,"litigationEvents":null,"insurancePostureDisclosed":null,"unitEconomicsDisclosed":null,"tickerSymbol":null,"stockExchange":null,"secCik":null,"revenueUsd":null,"revenueBasis":null,"revenueAsOf":null,"employeeCount":1204,"employeeCountBasis":"Wikipedia summary of company 10-K disclosure; more recent public number not located in reviewed materials","employeeCountAsOf":"2022-01-01T00:00:00.000Z","marketCapUsd":null,"marketCapAsOf":null,"marketCapSource":null,"createdAt":"2026-08-28T01:48:29.273Z","updatedAt":"2026-08-28T03:45:17.421Z","jsonLd":{"@context":"https://schema.org","@type":"Organization","@id":"https://registry.deploy.report/companies/digitalocean","url":"https://registry.deploy.report/companies/digitalocean","name":"DigitalOcean","description":"DigitalOcean is a publicly traded (NYSE: DOCN) developer and small-to-mid-market cloud that entered the GPU market in 2023 by acquiring Paperspace and folding its infrastructure into the DigitalOcean brand as \"GPU Droplets.\" Unlike the training-focused neocloud pack (CoreWeave, Crusoe, Nebius) chasing frontier-lab contracts on 100,000-GPU campuses, DigitalOcean sells GPU capacity the same way it sells VMs: per-second billing, self-serve via API, and single-GPU minimums aimed at startups, indie developers, and enterprises building inference or fine-tuning workloads rather than pre-training runs.\n\nIts GPU fleet is spread across colocation footprint (NYC2, TOR1, ATL1, RIC1, AMS3) rather than owned campuses; a March 2025 expansion deal with Flexential added high-density H200 and AMD Instinct capacity. The catalog covers Nvidia H100, H200, and Blackwell HGX B300 alongside AMD MI300X, MI325X, and MI350X, with on-demand pricing published between roughly $0.76 and $11.19 per GPU-hour. In April 2026 the company repositioned as an \"AI-native cloud built for the inference era\" and launched a managed Inference Engine, tying its GPU business to agentic and inference workloads rather than training scale.\n\nThe strategic bet is the long tail: whereas neoclouds compete on multi-hundred-megawatt anchor tenants, DigitalOcean's advantage is the same self-serve motion, catalog PCI, and roughly 600,000 paying customers that already power its core cloud, with GPU capacity offered as another SKU. Total company revenue was $692.9 million in 2023 (per Wikipedia summary of 10-K); AI-specific revenue is disclosed only qualitatively as a fast-growing segment inside overall results.","identifier":"4631991b-ab40-4993-ab2f-55b75714806f","foundingDate":"2011","address":"Broomfield, Colorado, USA (incorporated Delaware; historically NYC)","publisher":{"@id":"https://deploy.report/#organization"}},"framework_metadata":{"framework_schema_version":"0.1.0","verification_status":"verified","maturity_stage":null,"lifecycle_state":null,"architectural_position":{"cohort":null,"sub_cohorts":[]},"within_cohort_verified_vs_claimed_pair":null,"cap_flags":[],"verification_depth":{"sources_count":4,"primary_source_types":["knowledge-base","primary-company-ir"]}}}