generated: '2026-08-26' method: searched source: https://docs.plamo.preferredai.jp/en/console name: Preferred Networks PLaMo developer console and free surfaces description: >- Preferred Networks operates a real developer console for the PLaMo API and two free, no-cost ways to exercise PLaMo before or alongside paying for it. It does NOT publish a test/live key split, test fixtures, magic identifiers, time simulation, or any sandbox environment with separate semantics — those fields are recorded as absent rather than filled. console: published: true url: https://console.platform.preferredai.jp/ status: 200 manual: https://docs.plamo.preferredai.jp/en/console capabilities: - Create and manage a tenant, projects and members with three roles - Create, delete and re-generate personal and project API keys - View daily and monthly token usage and spend per project - Set per-project usage limits and alerts - Set a tenant monthly billing limit within the tenant quota free_surfaces: - name: PLaMo Chat (demo) url: https://chat-demo.plamo.preferredai.jp cost: free auth: email address only limits: 10 messages per minute; PLaMo 2.2 Prime only note: Browser chat UI, not an API surface. - name: Free plan status: announced-not-yet-available note: >- A zero-price API plan is published on the pricing page with the CTA disabled and labelled "in preparation"; PFN's 2026-06-22 plan notice states only Standard and Provider are available. See plans/preferred-networks-plans-pricing.yml. - name: Signup credit note: >- Campaign promotion codes grant yen-denominated credit at signup (e.g. an ICPC 2025 Yokohama grant of 20,000 yen valid to 2026-07-31). Campaign-specific and expiring, not a standing free tier. test_mode: separate_environment: false test_key_prefix: none published live_key_prefix: none published test_fixtures: none published magic_identifiers: none published time_simulation: none published note: >- There is one set of API keys and one set of endpoints. A developer rehearsing a call has no non-billing mode; the closest published affordance is POST /v1/tokenize, which prices the input side of a request without running a generation.