N8n shipped Gateway credits in version 2.36, letting developers on Cloud plans access AI model providers and automation tool services through a single shared balance instead of managing separate accounts and API keys for each.
The feature supports six model providers: OpenAI, Anthropic, Google Gemini, Alibaba Cloud's Qwen, Kimi.ai, and MiniMax. Five tool services round out the initial set: Brave Search, Browserbase, Firecrawl, LlamaParse, and PDF.co. Per n8n, adding a node for a supported provider now presents "Gateway credits" as an option alongside the standard credential selector, drawing usage from a prepaid balance tied to the n8n account rather than a separate provider subscription.
Pricing tracks each provider's published list rates, with all rates listed on a public page before any usage is charged. The Cloud Admin Panel logs spend by workflow and by service, supports manual top-ups, and includes automatic refill options. Instance owners can enable or disable Gateway credits globally from workspace settings.
The friction the feature removes is concrete. Before v2.36, building a workflow that called an unfamiliar AI provider required stopping to create an account, add a payment method, generate an API key, and configure a credential in n8n. For teams running workflows across multiple providers, that translates into parallel dashboards, separate invoices, and scattered expense reports. n8n says it targeted provider list pricing specifically to avoid marking up the convenience.
New self-service Cloud accounts and existing paying Starter and Pro customers received an initial credit allocation at rollout. Self-hosted instances are not included in the first release; n8n says extending the feature to self-hosted setups requires additional infrastructure work that is not part of v2.36.
Eleven providers and services are available at launch, with n8n saying the list will grow toward the APIs and capabilities teams use most, along with more granular per-instance controls for credit allocation.
