Gemini 3.6 Flash and GPT-5 Mini achieved perfect 100% results on our standing seven-day benchmark while recording zero spend over the testing period. Performance and cost data logged on October 7, 2026, across 13 models demonstrate clear operational trade-offs between speed, pricing, and benchmark scores. Across all monitored endpoints, uptime remained at 100% across hourly automated checks during the last 24 hours.
Benchmark
On the standing seven-day benchmark, Gemini 3.6 Flash and GPT-5 Mini reached the maximum available score of 100% with $0 total spend logged. GPT-5.6 Terra logged the second-highest score at 85.4%, closely followed by Gemini 3.1 Flash Lite at 84.8%. Further down the benchmark standings, Gemini 2.5 Flash recorded a 75.2% score, while GPT-5.6 Luna reached 74.4%. Every model tracked on the standing benchmark registered $0 in spend during this testing period.
| Model | 7-Day Benchmark Score | 7-Day Spend |
|---|---|---|
| Gemini 3.6 Flash | 100% | $0 |
| GPT-5 Mini | 100% | $0 |
| GPT-5.6 Terra | 85.4% | $0 |
| Gemini 3.1 Flash Lite | 84.8% | $0 |
| Gemini 2.5 Flash | 75.2% | $0 |
| GPT-5.6 Luna | 74.4% | $0 |
Speed
Latency measurements collected over the last 24 hours revealed significant variance across different model tiers. Gemini 2.5 Flash Lite recorded the fastest median response time at 596ms, remaining the only model to operate below the 600ms mark. OpenAI models followed closely, with GPT-5.4 Mini posting a 730ms median latency.
Mid-tier speed performance was anchored by Google's lite variants and sub-second models. Gemini 3.1 Flash Lite logged an 868ms median response time, while Gemini 2.5 Flash achieved 941ms. In the one-second range, OpenAI's GPT-5.6 series demonstrated consistent progression: GPT-5.6 Terra registered 1.11s, GPT-5.6 Luna logged 1.23s, and GPT-5.6 Sol recorded 1.31s median latency.
Larger or higher-capability models showed slower response times. Gemini 3.6 Flash reached 2.06s, followed by Gemini 3.5 Flash at 2.28s. GPT-5 Mini logged 2.36s, and GPT-5 Nano posted 2.39s median latency. The slowest measured endpoints were the Pro-tier offerings: Gemini 3.1 Pro (preview) recorded a median time of 3.54s, while Gemini 2.5 Pro came in at 3.78s. All 13 models maintained a 100% success rate across all hourly uptime checks.
Prices
Billed API pricing across the Gemini lineup spans a wide range based on input and output token volumes. Gemini 2.5 Flash Lite sits at the lowest cost tier, billed at $0.100 per million input tokens and $0.400 per million output tokens. Gemini 3.1 Flash Lite follows at $0.250 per million input tokens and $1.50 per million output tokens. Gemini 2.5 Flash is priced at $0.300 per million input and $2.50 per million output tokens.
At the higher tier, Gemini 2.5 Pro requires $1.25 per million input tokens and $10.00 per million output tokens. Gemini 3.5 Flash is billed at $1.50 per million input tokens and $9.00 per million output tokens. Gemini 3.1 Pro (preview) represents the highest input and output rate measured among the Google endpoints, costing $2.00 per million input tokens and $12.00 per million output tokens.
These measurements reflect direct operational results recorded under standard automated testing conditions over the specified time windows. They do not prove long-term infrastructure stability under custom enterprise workloads, nor do they guarantee identical latency under regional network variations or peak usage spikes.
