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: Shadeform
providerId: shadeform
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- GPU
- Cloud
- Marketplace
- Compute
- Infrastructure
- Rate Limiting
- Quotas
- Throttling
description: >-
  The Shadeform REST API authenticates every request with an X-API-KEY header.
  Specific published request-per-minute or request-per-day limits for the control
  plane are not documented in the public API reference. The more meaningful limits
  in a GPU marketplace are capacity-driven: real-time GPU availability per cloud,
  region, and GPU type (surfaced via GET /instances/types), and any per-account
  concurrent-instance or quota caps applied to a given account or tier. Specific
  values are not reconciled in this artifact.
notes: >-
  Confirm any documented control-plane request limits, per-account concurrent
  instance caps, and quota policies in the Shadeform dashboard or by contacting
  support during reconciliation. Effective deployability is bounded by live GPU
  availability rather than a fixed API call quota.
sources:
- https://docs.shadeform.ai/getting-started/quickstart
- https://docs.shadeform.ai/api-reference/instances/instances-types
responseCodes:
  throttled: 429
limits:
- name: Control-Plane Requests
  scope: account
  metric: requests
  limit: see provider documentation
  notes: No specific RPM/RPD published for the management API; verify in dashboard.
- name: GPU Availability
  scope: marketplace
  metric: capacity
  limit: real-time, per cloud / region / GPU type
  notes: Launch success depends on live availability; query GET /instances/types with available=true.
- name: Concurrent Instances
  scope: account
  metric: instances
  limit: see provider documentation
  notes: Per-account concurrent instance / quota caps may apply by tier; not reconciled.
policies:
- name: API Key Authentication
  description: Every request must carry a valid X-API-KEY header issued from the Shadeform dashboard.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
- name: Availability-Aware Retries
  description: When a launch fails for lack of capacity, re-query instance types/availability and retry against an available cloud, region, or GPU type.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com