Errors & limits

apisale enforces three independent limits:

1. Organization concurrency

Maximum parallel runs per organization (statuses: pending, assigned, processing).

Tier	Lifetime top-ups	Concurrent runs
starter	$0	10
growth	≥ $350	25
pro	≥ $1,400	50
scale	≥ $7,000	100

When exceeded:

{
  "error": {
    "code": "concurrency_limit_exceeded",
    "message": "Organization concurrency limit reached (10/10). Tier=starter.",
    "scope": "organization",
    "limit": 10,
    "in_flight": 10
  }
}

HTTP 429 Too Many Requests

2. API key concurrency

Each key has max_concurrency (default 10). A single key cannot exceed its own cap even if the org has headroom. scope: "api_key" in the error body.

3. Rate limit

Each key has rate_limit_per_minute (default 120). Exceeding it returns:

{
  "error": "rate_limit_exceeded",
  "message": "API key rate limit exceeded (120/minute)"
}

Client recommendations

Read X-Concurrency-Remaining on submit responses
Poll GET /v1/account for proactive backoff
Use exponential backoff on 429
Distribute load across multiple keys only within org policy

Insufficient balance

Submitting a run with insufficient available_usd returns 400 with an insufficient balance message. Top up in the Console.

​1. Organization concurrency

​2. API key concurrency

​3. Rate limit

​Client recommendations

​Insufficient balance

1. Organization concurrency

2. API key concurrency

3. Rate limit

Client recommendations

Insufficient balance