generated: '2026-07-19'
method: searched
source: https://docs.gradient.network/enterprise-solutions/gradient-cloud/api-reference-documentation
authentication:
  style: http-bearer
  header: Authorization
  format: Bearer <access-key>
  public_operations: [listModels]
  ref: authentication/gradient-authentication.yml
idempotency:
  supported: false
  notes: No idempotency-key header or contract is documented.
pagination:
  supported: false
  notes: No paginated collection endpoints are documented (model list returns a single object.list).
streaming:
  supported: true
  transport: server-sent-events
  content_type: text/event-stream
  request_field: stream
  terminator: 'data: [DONE]'
  applies_to: [createChatCompletion, createCompletion]
versioning:
  style: uri-path
  current: v1
  ref: lifecycle/gradient-lifecycle.yml
error_envelope:
  field: error
  shape: '{ "error": { "message": string, "type": string, "code": string } }'
  standard: openai-style
  ref: errors/gradient-problem-types.yml
rate_limiting:
  documented: true
  free_tier: 60 requests/minute, 3 concurrent
  paid_tier: plan-based, typically 1000-10000 requests/minute, 10-100 concurrent
  signal: 429 Rate Limit Exceeded
  retry_guidance: exponential backoff on 429/500/502/503/504 (per docs best-practices)
compatibility:
  standard: openai-compatible
  notes: Request/response objects mirror OpenAI; existing OpenAI client libraries work by pointing base_url at https://apis.gradient.network/api/v1.