Anthropic

AI

Claude chat completions (Sonnet, Opus, Haiku) via native and OpenAI-compatible APIs.

97.2
out of 100

High accuracy (93%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 672ms (p95: 1494ms).

Last tested: 3/19/2026, 5:41:50 PM

What AI models say

GPT-4o-mini

“Nine out of ten API requests succeeded, with high response quality and average response time of 796ms.”

Haiku

“9 of 10 requests succeeded with 93% response quality and 796ms 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.

Asking for a recommendation.

POST /v1/messagestypical672ms
100%

# The Midnight Library

Inquiring about the time.

POST /v1/messagestypical633ms
100%

I don't have access

Asking for the latest news.

POST /v1/messagestypical790ms
100%

I don't have access

Testing with an empty content string.

POST /v1/messagesadversarial300ms
30%

HTTP 400

Service Details

Endpoint: https://anthropic.mpp.tempo.xyz
Status: active
Discovered: 3/18/2026
Tests run: 10
Last tested: 3/19/2026

Test History

3/19/2026, 5:41:37 PM
score: 93%p50: 672ms
3/19/2026, 4:21:40 PM
score: 88%p50: 716ms
3/19/2026, 3:50:49 PM
score: 68%p50: 764ms
3/19/2026, 2:18:42 PM
score: 73%p50: 946ms
3/19/2026, 12:20:19 AM
score: 82%p50: 892ms

Embed this badge

MPPrimo rating for Anthropic<a href="https://mpprimo.com/service/40f7ac53-6951-4711-92e7-1036b780504e"><img src="https://mpprimo.com/api/badge/40f7ac53-6951-4711-92e7-1036b780504e" alt="MPPrimo rating"></a>
How do we test? Read our methodology.