The Cutoff ReportSearch

Cutoff Data

Model speed & uptime index

How fast each major model answers, and how often it fails, measured hourly from an independent client over the last 7 days.

Last reading 8/7/2026, 6:17:31 PM UTC

ModelProviderMedian95th pctUptimeProbesNow
Gemini 2.5 Flash Litegoogle/gemini-2.5-flash-liteGoogle846ms853ms100%2OK
GPT-5.6 Lunaopenai/gpt-5.6-lunaOpenAI1.06s1.25s100%2OK
Gemini 2.5 Flashgoogle/gemini-2.5-flashGoogle1.09s1.41s100%2OK
GPT-5.6 Solopenai/gpt-5.6-solOpenAI1.35s1.60s100%2OK
Gemini 3.1 Flash Litegoogle/gemini-3.1-flash-liteGoogle1.39s1.44s100%2OK
GPT-5.4 Miniopenai/gpt-5.4-miniOpenAI1.47s2.08s100%2OK
GPT-5.6 Terraopenai/gpt-5.6-terraOpenAI1.84s2.30s100%2OK
Gemini 3.5 Flashgoogle/gemini-3.5-flashGoogle2.13s2.43s100%2OK
Gemini 3.6 Flashgoogle/gemini-3.6-flashGoogle2.57s2.87s100%2OK
GPT-5 Nanoopenai/gpt-5-nanoOpenAI2.81s3.15s100%2OK
GPT-5 Miniopenai/gpt-5-miniOpenAI3.05s3.33s100%2OK
Gemini 2.5 Progoogle/gemini-2.5-proGoogle4.00s4.21s100%2OK
Gemini 3.1 Pro (preview)google/gemini-3.1-pro-previewGoogle4.08s4.17s100%2OK

How this is measured

Once an hour we send every tracked model the same short, deterministic prompt and record wall-clock time from request to completed response, plus whether the call succeeded.

Median and 95th-percentile figures are computed across all successful probes in the trailing 7-day window. Uptime is successful probes divided by total probes.

These are single-client measurements over one network route, at short prompt lengths. They are a fair basis for comparing models against each other under identical conditions, not a substitute for a provider’s own status page.