Are there mock/test implementations of the APIs that i can use that are not subject to the same rate limits as the prod APIs? For exploratory programming or running tests against. Data persistence is not necessary for this use case.
Hi Joonas L., Please stand by as I search the Mixpanel knowledge base for an AI-powered answer. I normally take about 15 seconds or so. In case AI cannot answer your question, feel free to reach out to Mixpanel support at mixpanel.com/get-support. If you have follow-up questions for this thread, please tag me again with your question: @kapa.ai <follow-up question> 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
The documentation you’ve shared only describes rate limits for the real (production) Mixpanel APIs and does not mention any mock, sandbox, or special test endpoints that are exempt from those limits. [Rate limits]
What is recommended for testing is:
- Use a separate development project vs. your production project so you can experiment safely, but this still uses the normal APIs and limits. [Check implementation] - For high-volume ingestion or backfill, follow the documented patterns (batching, concurrency, exponential backoff on 429s) rather than relying on a special test API. [Import events rate limit; Track events]
I’m sorry, but there’s no information in these sources about any official mock/test implementations of Mixpanel APIs that bypass rate limits.
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
kapa.ai I'm already using a separate test project. I have very tiny amounts of data i use for testing. I only need data to stick around until i complete a test run, after that it can all dissappear. I need a solution that can sustan more load than 60 queries an hour, otherwise my project is effectively un-testable.
Let me check the knowledge base for you again. One moment! 📚
Mixpanel does not document any special mock, sandbox, or “test-only” API endpoints that bypass rate limits. All of the following use the same global limits:
- Query API (Insights, Funnels, Segmentation, JQL, etc.): 5 concurrent queries, 60 queries per hour. [Rate limits; JQL] - Raw Data Export API: 100 concurrent, 60 per hour, 3 per second. [Rate limits; Raw export] - Ingestion API: very high throughput (2GB/min or ~30k events/sec), but that doesn’t help with Query API limits for testing queries. [Import rate limit]
The only documented mitigations for rate‑limiting are:
- Spread queries out over time. - Consolidate multiple queries into a single query where possible (e.g., use one segmentation query instead of many similar ones). [Rate limits]
Using a separate dev/test project is recommended to protect production data quality, but it does not change rate limits. [Separate dev data]
Based on the provided docs, there is no supported way to:
- Temporarily disable rate limits, - Use a special “ephemeral data” or “test” API, - Or otherwise exceed 60 Query API calls per hour for automated/exploratory tests.
To go beyond that for your test use case, the only path suggested in the docs is to adjust how many queries you run (batching/consolidation) and their scheduling; anything beyond that would require contacting Mixpanel (e.g., support/CSM), which is not documented in detail here.
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
