lightningai-rate-limit.yaml 768 B

123456789101112131415161718192021
  1. name: lightningai-rate-limit
  2. repo: https://github.com/dinosauria123/hermes-lightningai-rate-limit
  3. sha: dc7911a946f0b209a81cf680c168382177bb13bf
  4. description: >-
  5. Per-process throttle for Lightning AI (https://lightning.ai/api/v1) keeping LLM requests
  6. within the ~15 requests/min limit, using an in-memory token bucket. Gates only calls whose
  7. provider or base_url points at Lightning AI; no-op otherwise.
  8. Disclosure — registers one llm_execution middleware and nothing else; no network calls,
  9. files, or processes of its own.
  10. maintainer: dinosauria123
  11. tier: community
  12. category: models
  13. requires_hermes: ">=0.21.5"
  14. version: "1.1.0"
  15. platforms: []
  16. capabilities:
  17. provides_tools: []
  18. provides_hooks: []
  19. provides_middleware:
  20. - llm_execution
  21. requires_env: []