specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: glhf
providerId: glhf-chat
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- AI
- LLM
- Inference
- Open Source Models
- Hugging Face
- Rate Limiting
- Quotas
- Throttling
description: >-
  glhf serves an OpenAI-compatible API backed by an auto-scaling GPU scheduler.
  Public materials note that the service initially launched without rate
  limiting and that limits have since been applied, but specific per-account or
  per-model request and token limits are not publicly documented. As an
  OpenAI-compatible surface, throttling is expected to be returned as HTTP 429.
  Specific values are not reconciled in this artifact.
notes: >-
  Verify any RPM, TPM, concurrency, or per-model limits in the glhf account
  settings and documentation during reconciliation; values are not currently
  published.
sources:
- https://glhf.chat
- https://glhf.chat/users/settings/api
responseCodes:
  throttled: 429
limits:
- name: Requests Per Minute (RPM)
  scope: account
  metric: requests
  limit: see provider documentation
  notes: Not publicly documented; limits applied after initial launch.
- name: Tokens Per Minute (TPM)
  scope: account
  metric: tokens
  limit: see provider documentation
  notes: Not publicly documented.
- name: Concurrent Requests
  scope: account
  metric: requests
  limit: see provider documentation
  notes: Governed by the auto-scaling GPU scheduler; not publicly documented.
policies:
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com