Blink
Blink uses 28 subprocessors and infrastructure vendors, including AWS, Google Cloud and OpenAI, based on its public subprocessor list. Last checked Nov 13, 2026.
AI Stack
Web Search
Foundation Models
LLM inference (API-tier — no training on inputs)
Isolated sandboxes that build and preview customer projects, file storage for customer apps, and inference for some AI models (Vertex AI)
LLM inference (API-tier — no training on inputs)
LLM inference (API-tier — no training on inputs)
Image & Video Generation
Infrastructure
Cloud & Networking
AWSPrimary Postgres database (Amazon Aurora), API servers that customer apps call, background jobs and scheduled tasks (ECS), static hosting for customer-deployed applications (S3, CloudFront, Lambda), project version history, email sent from customer domains (SES), and inference for some AI models (Bedrock)
CDN, DNS, edge caching, Workers (Pro+ customer backends), KV/R2 storage
Per-project AI agent sandboxes (Blink Claw)
Isolated sandboxes that build and preview customer projects, file storage for customer apps, and inference for some AI models (Vertex AI)
No live customer traffic. Stopped copies of earlier databases and caches remain on disk until they are deleted, and some internal staff tools still run there
Domain registration, including registrant contact details
Application hosting, serverless functions, durable workflow execution, and the AI gateway that routes model requests to the LLM providers below
Tunnels for previewing mobile apps on a device
Databases
Billing & Payments
Analytics & BI
Communications
Model labs reached through vendors 4
Blink does not list these model providers. Vendors on its list name them on their own subprocessor lists.
Changes
Checked Nov 13, 2026Source
This is the first saved copy of Blink’s subprocessor list. Added and removed vendors will show here after the next check.
Similar companies
Companies with the most overlapping vendors