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: Thunder Compute
providerId: thundercompute
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- GPU
- Cloud
- Infrastructure
- AI
- Compute
- Rate Limiting
- Quotas
- Throttling
description: >-
  Thunder Compute does not publish explicit numeric request-per-minute rate
  limits for its REST API. The primary constraints are account-level resource
  quotas - the number and type of GPU instances an account may provision
  concurrently - which are governed by available GPU capacity, the account's
  tier/standing, and payment status rather than by API call throttling. New
  accounts may have lower concurrent-instance and GPU-type quotas that are
  raised on request. Specific numeric limits are not reconciled in this
  artifact.
notes: >-
  No public rate-limit documentation page exists as of 2026-06-21. Verify
  concurrent-instance and GPU quotas in the Thunder Compute console or with
  support; honor HTTP 429 responses with backoff if returned.
sources:
- https://www.thundercompute.com/docs
- https://www.thundercompute.com/pricing
responseCodes:
  throttled: 429
limits:
- name: Concurrent Instances
  scope: account
  metric: instances
  limit: see provider console
  notes: Number of simultaneously running GPU instances; governed by quota and available capacity.
- name: GPU Type Availability
  scope: account
  metric: gpus
  limit: see provider console
  notes: Access to specific GPU models (e.g. H100, A100) may be gated by account tier and inventory.
- name: API Requests
  scope: account
  metric: requests
  limit: not published
  notes: No documented per-minute REST request cap; clients should still handle 429 with backoff.
policies:
- name: Quota-Based Limits
  description: Constraints are primarily concurrent-instance and GPU-type quotas, raised on request as accounts establish history and spend.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After if a 429 is returned.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com