OpenRouter
AIUnified access to 100+ LLMs with single API.
97.2
out of 100
High accuracy (93%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 655ms (p95: 1372ms).
Last tested: 3/19/2026, 5:42:43 PM
What AI models say
GPT-4o-mini
“Completed 10 requests with 9 successes, 1 error, 93% response quality, and an average response time of 744ms.”
Haiku
“9 of 10 requests succeeded with 93% response quality and 744ms average latency.”
Accuracy (40%)
93.0
Reliability (30%)
100.0
Latency (20%)
100.0
Cost Efficiency (10%)
100.0
Test Examples
Real requests we sent and the responses we received.
User input with special characters.
POST /v1/chat/completionsedge587ms100%
It looks like you've typed
User inquiring about a programming concept.
POST /v1/chat/completionstypical949ms100%
Inheritance is a fundamental concept
User requesting a summary of a topic.
POST /v1/chat/completionstypical559ms100%
Certainly! The theory of
User providing an empty content string.
POST /v1/chat/completionsadversarial397ms30%
HTTP 400
Service Details
Endpoint:
https://openrouter.mpp.tempo.xyzStatus: active
Discovered: 3/18/2026
Tests run: 10
Last tested: 3/19/2026
Test History
3/19/2026, 5:42:30 PM
score: 93%p50: 655ms
3/19/2026, 4:22:28 PM
score: 83%p50: 604ms
3/19/2026, 3:51:42 PM
score: 84%p50: 1539ms
3/19/2026, 12:21:10 AM
score: 84%p50: 643ms
Embed this badge
<a href="https://mpprimo.com/service/27e316e0-5697-4c00-8188-e1205ae3227a"><img src="https://mpprimo.com/api/badge/27e316e0-5697-4c00-8188-e1205ae3227a" alt="MPPrimo rating"></a>