generated: '2026-08-17' method: searched source: https://www.edgee.ai/pricing currency: USD plan_count: 3 note: 'Seat-priced, not call-priced: Edgee charges per developer per month for the gateway and takes a platform fee on credit top-ups, while the underlying model tokens are either billed through Edgee credits or through the customer''s own provider keys (BYOK). There is no published per-request or per-token rate card for the gateway itself.' plans: - name: Free price: 0 currency: USD period: month unit: developer included: - 1 developer seat - Multi-agent support (Claude Code, Codex, Copilot and others) - Token compression (input and output) - Personal dashboard with usage, cost and session tracking limits: seats: 1 scope: single developer only overage: null - name: Team price: 29 currency: USD period: month unit: developer billing: per developer per month included: - Everything in Free - BYOK plus a daily allowance on Edgee-hosted models - Fallback and reroute models - Team observability and squad management - Spending cap per seat with alerts - GitHub integration with per-repo and per-PR attribution limits: spend_cap: per seat, with alerts overage: platform_fee: 5% applies_to: Edgee credit top-ups - name: Enterprise price: null price_model: custom currency: USD period: custom contact_required: true included: - Everything in Team - Private open-weight models, billed per GPU hour - SSO / SAML authentication - Private gateway (SaaS or on-premises) - Custom data residency (EU or US) - Custom privacy controls, including PII redaction - 30-minute SLA on Urgent tickets, 24/7 web and phone support limits: tokens: unlimited on private models overage: model: custom pricing self_hosting: available: true source: https://www.edgee.ai/docs/self-hosting/overview modes: [docker-compose, kubernetes-helm] note: 'The AI Gateway can be run on the customer''s own infrastructure, connected or air-gapped (headless). Pricing for self-hosted deployment is not published; it sits under Enterprise.'