X-RateLimit-Limit
The short-window request limit for the current key and endpoint.
YouTube API Quota and Rate Limits
Understand YouTube API quota, transcript API rate limits, monthly request limits, and how to handle quota_exceeded and rate_limited errors.
Direct answer
Quota and rate limits are different controls
Quota usually means a monthly or account-level usage allowance. Rate limit means a short-window protection against bursts. Scority exposes both concepts through distinct error codes.
- quota_exceeded means the API key reached its monthly product quota.
- rate_limited means the integration sent too many requests in a short period.
- Both return HTTP 429, but your app should handle them differently.
Concept
YouTube Data API quota concept
The official YouTube Data API uses Google's own quota model. If you are building against official Google endpoints, check the current Google documentation for quota units and billing behavior before publishing exact numbers.
- Official Google quota is separate from Scority quota.
- A transcript-specific API may have its own monthly quota and short-window limits.
- Do not mix Google quota unit claims with Scority usage claims.
Scority quota
Scority monthly quota
Scority API keys can have a monthly quota. When the key has no remaining monthly allowance, the API returns quota_exceeded.
- Monthly quota is tracked per API key/account setup.
- Usage reports are visible in the dashboard foundation for linked accounts.
- Quota changes are handled through the operator/support flow unless automated billing is enabled later.
Scority limits
Scority short-window rate limit
Short-window rate limiting protects service reliability. It is designed to slow bursts, not to replace monthly quota.
- Batch jobs should queue requests instead of firing large bursts.
- Integrations should read Retry-After when present.
- Do not retry rate_limited responses immediately.
Rate-limit headers
Use these headers to slow down before a burst turns into user-facing errors.
X-RateLimit-Remaining
How many requests remain in the current short-window interval.
X-RateLimit-Reset
When the current short-window interval resets.
Retry-After
How many seconds to wait before retrying after a 429 rate_limited response.
Errors
quota_exceeded
quota_exceeded means the API key has reached its monthly product quota. Retrying immediately usually does not help because the monthly allowance has not changed.
- Show a clear account or operator message.
- Reduce usage, wait for reset or request a quota change.
- Check whether repeated background jobs are consuming more quota than expected.
Errors
rate_limited
rate_limited means the integration crossed a short-window request limit. This is usually fixed by slowing down, queuing work or retrying after the indicated delay.
- Respect Retry-After when present.
- Use backoff instead of immediate loops.
- Avoid running many transcript requests at page-load time.
Reduce pressure
Caching and retries
Most transcript workflows can reduce quota and rate-limit pressure by avoiding repeated fresh fetches for the same video and language.
- Cache successful transcript responses in your own app when appropriate.
- Avoid retrying invalid_video_id or invalid_video_url requests.
- Retry upstream_transcript_failed later rather than in a tight loop.
- Separate planned bulk jobs from interactive user requests.
Docs
Rate limits and quotas
See the canonical rate-limit header reference.
Open →Quota
Quota exceeded guide
Troubleshoot monthly quota exhaustion and usage planning.
Open →429
429 error guide
Troubleshoot rate_limited, quota_exceeded and retry behavior.
Open →Errors
Error codes
Review quota_exceeded, rate_limited and retry guidance.
Open →Pricing
Pricing concepts
Understand cost drivers for transcript workflows.
Open →