Rate Limits - VLM Run

The VLM Run Gateway applies per-tier rate limits on POST /v1/openai/chat/completions. Limits are enforced in memory; when you exceed a bucket, the API returns 429 Too Many Requests. See Authentication for how tiers are determined.

Limits by Tier

Tier	How it is identified	Rate limit	Bucket key
Anonymous	No `Authorization` header, or `Bearer ""` / `Bearer vlmrun`	120 requests/min, 1000 requests/hr	Client IP
Authenticated	Valid VLM Run API key (`Authorization: Bearer <VLMRUN_API_KEY>`)	240 requests/min, 10000 requests/hr	User ID

Authenticated users get a higher quota (240/min · 10000/hr) than anonymous traffic (120/min · 1000/hr). Attribution also differs: anonymous requests are capped per IP, while authenticated requests are capped per user account. See Authentication for how the Gateway resolves tiers from the Authorization header.

Sign up at app.vlm.run for a free API key. An API key lets you attribute usage to your account and avoids sharing a rate limit bucket with other users on the same IP.

When a limit is exceeded, the Gateway returns HTTP 429. See Error Codes for the response body and handling guidance.

Tips

Back off and retry after a short delay when you receive a 429.
Prefer URL-based document inputs over large base64 payloads to keep request latency predictable under load.
For production document workloads, use an authenticated API key so limits apply to your account rather than a shared IP bucket.

If you need higher limits, contact support@vlm.run.