Head to head
- KimiCold end-to-end latency · lower is betterGitHub faster4.24x
GitHub
499msLLM API
2.12s
Kimi
Median cold end-to-end latency over time
GitHubLLM API
Work at GitHub or LLM API? See where you stand every day, with full run logs, not just on benchmark Friday.Get stealth benchmarks →
Methodology
- One workload, every provider. The same script, payload, and model hit every provider, addressed the way its own API expects. Nobody gets a tuned or special path.
- Repeated on a schedule. Every suite runs many iterations per provider, automated on a recurring schedule, so results reflect steady state rather than one lucky request.
- Raw results published. Every run is committed as JSON to the public benchmarks repo, next to the code that produced it. Nothing is cherry-picked or held back.
- How this comparison is built. Each benchmark compares the two providers' medians; results within 5% count as roughly even. A comparison is published only when both providers completed the latency benchmark on at least two of the same model families reliably: at least 50% success for each provider, in benchmark runs from the last 30 days. A pair also qualifies when both providers completed at least 90% of runs in one shared benchmark. If a provider’s latest run of a benchmark failed, its most recent successful run from that window is used instead.
Frequently asked questions
Have another question? Email us.
GitHub was faster in the only shared AI gateway benchmark. In Kimi (October 30, 2026), GitHub had a median cold end-to-end latency of 499ms versus 2.12s for LLM API, 4.24x faster.