Rate limits

Last updated Jun 14, 2026

Matter applies rate limits per key or token. The defaults are generous for normal use and scale on request.

Default limits

Agent tokens can be granted burst exemptions via the limits.frequency_caps field on the token's structured policy. This is how tier-4 automation agents avoid rate-limit contention during scheduled filing runs.

Rate-limit headers on every response

X-RateLimit-Limit: 100
X-RateLimit-Remaining: 99
X-RateLimit-Reset: 1745251260

Budget your request rate against these. X-RateLimit-Reset is a Unix timestamp.

429 response shape

When you exceed a limit, Matter returns a 7807 problem object plus the standard Retry-After header (seconds):

HTTP/1.1 429 Too Many Requests
Retry-After: 30
X-RateLimit-Limit: 100
X-RateLimit-Remaining: 0
X-RateLimit-Reset: 1745251290
Content-Type: application/problem+json

{
  "type": "https://mattermode.com/docs/errors/rate_limited",
  "title": "Too Many Requests",
  "status": 429,
  "code": "rate_limited",
  "detail": "This key exceeded the 100 req/s burst limit. Retry after 30s.",
  "retry_after": 30,
  "request_id": "req_Qw9xYz8A"
}

Retry strategy

Respect Retry-After. If it's absent on a transient 5xx, back off exponentially starting at 1s with jitter: 1s, 2s, 4s, 8s, 16s — stop after 5 attempts for a single logical request.

Most SDK clients do this automatically. If you're writing against the API directly, use a library like backoff (Python) or p-retry (JavaScript).

Raising limits

Send support@mattermode.com with:

• Account ID.
• Typical requests per second at peak.
• Burst-rate needs (if any).
• Use case (compliance automation, venture studio batching, etc.).

Production accounts with established traffic patterns are routinely raised to 500+ RPS burst. Venture-studio accounts running portfolio batching workloads use a separate burst budget tied to the Portfolio resource.