hashference-provider.yaml 1.2 KB

12345678910111213141516171819
  1. name: hashference-provider
  2. title: Hashference
  3. repo: https://github.com/0xalank/hermes-hashference-provider
  4. sha: 82700f5377a46f0e7bd482c907ed04090b6305be
  5. subdir: hashference-provider
  6. version: "1.0.0"
  7. description: "Hashference as a Hermes model provider (`--provider hashference`, or Hashference in `hermes model`): open and frontier models on one balance, funded by mining, QUAI, Qi or staking. Each model's real limits come from Hashference's catalog: the hosted models accept 98,304 input tokens inside a 163,840-token window and cap output at 8,192, so Hermes compresses before an oversized-input refusal; models that cannot call tools are left out of the model list. Disclosure — model provider; chat requests (full prompts) go to https://hashference.com/v1 (or your configured base_url) using HASHFERENCE_API_KEY read by Hermes; the plugin itself only sends an unauthenticated GET of the public model catalog (/v1/models), cached ten minutes."
  8. maintainer: 0xalank
  9. tier: community
  10. category: models
  11. requires_hermes: ">=0.22.5"
  12. docs_url: https://github.com/0xalank/hermes-hashference-provider
  13. platforms: []
  14. capabilities:
  15. provides_tools: []
  16. provides_hooks: []
  17. provides_middleware: []
  18. requires_env:
  19. - HASHFERENCE_API_KEY