Based on the knowledge sources provided, I can explain the difference between rate limits for batch check permissions endpoint versus the regular check permissions endpoint in Ory. ## Rate Limits Ory Network applies rate limits based on your subscription plan and project environment. For permission checks specifically: 1. Single Permission Check (

/relation-tuples/check

): - This endpoint has specific rate limits depending on your plan - For example, in the Developer plan, it's limited to 10 requests per second (burst) and 300 requests per minute (sustained) - In the Production plan for production environments, it's 80 requests per second and 1800 requests per minute - In the Growth plan for production environments, it's 800 requests per second and 18000 requests per minute [Understand Ory Network rate limiting](https://www.ory.sh/docs/guides/rate-limits#project-rate-limits-in-workspaces) 2. Batch Permission Check (

/relation-tuples/batch/check

): - The batch endpoint allows you to check multiple permissions in a single API call - This is more efficient when you need to check multiple permissions at once - The documentation doesn't specify different rate limits specifically for the batch endpoint The main advantage of using the batch endpoint is efficiency - you can check multiple permissions in a single API call, which helps you stay within your rate limits by reducing the number of separate API calls needed. ## Latency The knowledge sources don't provide specific information about latency differences between the single check and batch check endpoints. However, generally, batch operations are more efficient in terms of network overhead since they reduce the number of HTTP requests needed. It's worth noting that there's an open GitHub issue (#1626) mentioning some inconsistencies between different permission checking methods, but it doesn't provide details about latency differences. If you need more specific information about latency differences, you might want to conduct your own performance testing or contact Ory support directly.

Based on the knowledge sources provided, there isn't explicit information about specific limits on the number of user identities or relationships that can be stored in Ory Cloud across different plans. The pricing and plan information focuses primarily on: 1. Rate limits for API calls (such as identity retrieval) 2. Number of environments allowed per plan 3. Features available in each plan 4. Pricing structure based on usage metrics like Average Daily Active Users (aDAUs), Machine-to-Machine Tokens, and Permission Checks The rate limits vary by plan (Developer, Production, Growth, Enterprise) and environment type (Production, Staging, Development), but these limits apply to API requests rather than storage capacity for identities or relationships [Understand Ory Network rate limiting](https://www.ory.sh/docs/guides/rate-limits#project-rate-limits-in-workspaces). For example, the

GET /admin/identities

endpoint has the following rate limits: - Developer plan: 1 request per second, 10 requests per minute - Production plan (production environment): 10 requests per second, 300 requests per minute - Growth plan (production environment): 20 requests per second, 600 requests per minute - Enterprise plan (production environment): 60 requests per second, 1200 requests per minute There are also some relevant GitHub issues discussing identity retrieval limitations and batch operations, but they don't mention storage limits for the number of identities or relationships: 1. Issue #305 discusses the need for retrieving subsets of identities through admin API [GitHub Issue #305](https://github.com/ory/network/issues/305) 2. Issue #396 mentions challenges with querying identity properties due to rate limiting [GitHub Issue #396](https://github.com/ory/network/issues/396) If you need specific information about storage limits for identities and relationships in Ory Cloud, I would recommend contacting Ory directly as this information is not explicitly stated in the provided knowledge sources.