Platform
Objective provides a platform of APIs offering search, finetuning, evaluations and more. All of our APIs are backed by the same hardened infra, serving hundreds of millions of requests per month. Read below to learn more about the reliability of our APIs.
Search latency
Our end to end search latency averages around 70ms per request. Our infrastructure scales elastically to support load based on user demand. All plans include a max of 1 query per second. For use cases requiring higher QPS, get in touch.
Search latency:
| Metric | Value |
|---|---|
| p50 | 70ms |
| p95 | 100ms |
Check out a live demo load testing our APIs:
Indexing
Indexing is event driven, occuring in near real time based on updates to the data. Our indexing system is extremely high throughput, handling thousands of updates per second daily at above 99.9% reliability. We deliver these updates quickly, with the average data update taking 3 seconds to reflect in the index.
Indexing end to end latency:
| Metric | Value |
|---|---|
| p50 | 3s |
| p95 | 18s |
Supported geographic regions
The following regions are available based on your plan:
| Region | Plan |
|---|---|
| us-east | Starter, Pro, Enterprise |
| us-west | Enterprise |
| eu-west | Enterprise |
| eu-central | Enterprise |
| ap-northeast | Enterprise |
| ap-southeast | Enterprise |
| sa-east | Enterprise |
Your region not available? Get in touch.