generated: '2026-09-03'
method: searched
source: https://imaginode.ai/en/docs/api-mcp
limit_count: 1
limits:
  - scope: per-api-key
    surface: POST /api/generate and MCP generation tools (REST API and MCP server share the limit)
    limit: 10 generations
    window: 1 minute
    burst: not documented
    note: >-
      Stated in the API docs as a deliberate runaway-agent guard: "each key is limited to
      10 generations per minute: plenty for real work, not enough to drain a balance by
      accident if an agent runs away." An account may hold at most 5 active API keys.
response_headers: []
response_headers_note: >-
  No rate-limit response headers (X-RateLimit-*, RateLimit-*, Retry-After) are documented,
  and the status code returned on exhaustion is not documented.
exhaustion_status: not documented