Rate limits

Plan	Applies to	Minute limit	Daily limit	Keyed by
Free	`/api/v1/public/*`	50 requests / 10 seconds	—	Client IP at the edge
Pro	Authenticated realtime API	30 requests/minute	20,000/day	Organization
Premium	Authenticated realtime API	180 requests/minute	150,000/day	Organization

Paid limits are organization-wide. Every API key and authenticated session in an organization draws from the same budgets. Creating more keys does not increase throughput. The daily budget resets at 00:00 UTC.

Public requests are limited and cached at the Cloudflare edge and are not double-counted by the application limiter. An IP over the public budget receives 429 Too Many Requests:

1234{
  "error": "too_many_requests",
  "message": "Public API rate limit exceeded. Retry after 10s."
}

Wait 10 seconds, then retry. Do not retry immediately; that extends the block.

Realtime reads from an organization with no paid plan are refused with 403 (see No plan at all). They are not billed against a minute or daily budget, so they carry none of the headers below.

Response headers

Paid (Pro and Premium) responses carry both budgets:

Header	Meaning
`RateLimit-Limit`	Plan requests per minute.
`RateLimit-Remaining`	Requests left in the current minute.
`RateLimit-Reset`	Seconds until the minute counter resets.
`RateLimit-Daily-Limit`	Plan requests per UTC day.
`RateLimit-Daily-Remaining`	Requests left today.
`RateLimit-Daily-Reset`	Seconds until 00:00 UTC.

An over-limit request returns 429 Too Many Requests. It includes the same headers, sets the exhausted remaining value to zero, and includes Retry-After for the budget that blocked the request:

123456{
  "error": "rate_limited",
  "limit": "rpm",
  "message": "Organization rpm request limit exceeded; retry after 18s.",
  "upgrade_url": "https://adjacent.markets/subscribe"
}

Back off for at least Retry-After seconds. upgrade_url points to the plan catalog. It is omitted when no plan is available to buy.

The live minute and daily meters are under Settings → Billing & usage.

Concurrent requests

Beyond the per-minute and daily budgets, each organization can run up to 4 requests at the same time. This capacity is dedicated to your organization, not shared with other callers. A request over the limit waits up to 2 seconds for a slot. If no slot frees up, it returns 429 Too Many Requests with Retry-After and the body above with "limit": "concurrency".

The public tier (/api/v1/public/*) shares one pool of capacity across all anonymous callers and can return 429 with Retry-After under load:

1234{
  "error": "too_many_requests",
  "message": "The public API is at its concurrent request limit. Retry after 2s."
}

Issue requests sequentially, or keep parallelism within the limit, and back off for Retry-After seconds on a concurrency 429.

No plan at all

A 429 means a plan's budget ran out. A request from an organization with no paid plan is refused up front instead, with 403 Forbidden:

12345{
  "error": "Insufficient scope: a Pro or Premium API plan is required",
  "status": 403,
  "upgrade_url": "https://adjacent.markets/subscribe"
}

upgrade_url carries the same meaning as on a 429 and is likewise omitted when no plan is available to buy. Unlike a 429, retrying does not help: nothing resets until the organization holds a plan.