LLM API

AI Gateway

LLM APIAI Gateway · public evidence

4 published workloads · 4 measured

Latest measured result

Nov 13, 2026, 09:32 UTC

Each workload retains its own metric and measurement date. No category-wide composite is published for this offering.

Published results for LLM API.

Compare workloads separately. Failed, skipped and incomplete results are not measured zeroes.

Public links open the latest workload results and methodology, not an immutable archive of the dates shown.

Median cold end-to-end latency · Lower is better

Nov 13, 2026, 09:32 UTCPublic source

Result2,309.26 ms

Rank

Median cold end-to-end latency · Lower is better

Nov 13, 2026, 09:07 UTCPublic source

Result1,077.74 ms

Rank#7of 8

Median cold end-to-end latency · Lower is better

Nov 13, 2026, 09:07 UTCPublic source

Result1,844.55 ms

Rank#4of 6

Median cold end-to-end latency · Lower is better

Nov 13, 2026, 09:05 UTCPublic source

Result700.87 ms

Rank#7of 7