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: Agenta
providerId: agenta
created: '2026-06-20'
modified: '2026-06-20'
reconciled: false
tags:
- AI
- LLMOps
- Prompt Management
- LLM Evaluation
- Observability
- Rate Limiting
- Quotas
- Throttling
description: >-
  Agenta does not publish explicit numeric per-endpoint request rate limits for
  its cloud REST API. Practical usage is bounded by plan quotas - included
  monthly traces (5k Hobby, 10k Pro, 1M Business) with $5 per additional 10k
  trace overage, monthly evaluation allowances (20 on Hobby, unlimited on Pro
  and above), seat counts, and data retention windows - rather than by RPM/TPM
  throttling. Self-hosted deployments are bounded only by the operator's own
  infrastructure. Specific per-endpoint throttling values are not reconciled in
  this artifact.
notes: >-
  Verify any documented per-endpoint throttling, 429 behavior, and OTLP
  ingestion limits in the Agenta docs and dashboard on reconciliation; quota
  allowances scale with the cloud plan and self-hosting removes platform quotas.
sources:
- https://agenta.ai/pricing
- https://docs.agenta.ai/
- https://docs.agenta.ai/observability/overview
responseCodes:
  throttled: 429
limits:
- name: Included Traces
  scope: account
  metric: traces
  limit: plan-based (5k Hobby / 10k Pro / 1M Business per month)
  notes: Observability traces included per month; overage billed at $5 per additional 10k on paid plans.
- name: Included Evaluations
  scope: account
  metric: evaluations
  limit: 20/month Hobby; unlimited Pro and above
  notes: Monthly evaluation runs allowance; varies by plan.
- name: Seats
  scope: account
  metric: seats
  limit: 2 Hobby / up to 10 Pro / unlimited Business
  notes: Workspace member seats; not a request rate limit.
- name: Data Retention
  scope: account
  metric: days
  limit: 30 Hobby / 90 Pro / 365 Business
  notes: Trace and evaluation data retention window per plan.
- name: Per-Endpoint Request Rate
  scope: account
  metric: requests
  limit: see provider documentation
  notes: No explicit numeric RPM/TPM published for the REST or OTLP endpoints as of the review date.
policies:
- name: Quota-Based Usage
  description: Cloud usage is bounded by plan quotas (traces, evaluations, seats, retention) rather than published RPM throttling.
- name: Self-Hosting Removes Platform Quotas
  description: MIT-licensed self-hosted deployments are limited only by the operator's own infrastructure and LLM provider limits.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor any Retry-After header on 429 responses.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com