Gemini 3.6 Flash and GPT-5 Mini achieved perfect 100% scores on the 7-day standing benchmark evaluation, leading all evaluated models while incurring $0 in baseline testing spend. Across 24 hours of performance tracking, [Gemini 2.5 Flash Lite](/models/google-gemini-2-5-flash-lite) delivered the fastest response speed, logging a 519ms median latency with 100% uptime across all checks.
Benchmark
In the 7-day standing benchmark evaluation, two models captured maximum accuracy. Gemini 3.6 Flash and GPT-5 Mini each reached 100% of available points while recording $0 in total spend during the evaluation period.
Behind the top performers, GPT-5.6 Terra secured 88.9% of available points at $0 total spend. Gemini 3.1 Flash Lite recorded an 86% benchmark score, followed by Gemini 2.5 Flash at 80.2% and GPT-5.6 Luna at 75.8%. All models evaluated in the benchmark suite logged zero total spend over the 7-day testing window.
| Model | Benchmark Score | Total Spend |
|---|---|---|
| Gemini 3.6 Flash | 100% | $0 |
| GPT-5 Mini | 100% | $0 |
| GPT-5.6 Terra | 88.9% | $0 |
| Gemini 3.1 Flash Lite | 86% | $0 |
| Gemini 2.5 Flash | 80.2% | $0 |
| GPT-5.6 Luna | 75.8% | $0 |
Speed
Response times gathered over the last 24 hours across hourly check-ins showed 100% of checks succeeded for all 13 tracked models. Gemini 2.5 Flash Lite registered the fastest response time overall, logging a 519ms median latency.
GPT-5.4 Mini was the second fastest model measured, yielding a 724ms median response time. Gemini 3.1 Flash Lite recorded 844ms, closely followed by Gemini 2.5 Flash at 869ms. Three variants from the GPT-5.6 series delivered median latencies between 1.1 seconds and 1.4 seconds: GPT-5.6 Luna at 1.13s, GPT-5.6 Sol at 1.30s, and GPT-5.6 Terra at 1.35s.
Slower response times were recorded across higher-tier and newer Flash offerings. Gemini 3.5 Flash checked in at 2.27s median latency, Gemini 3.6 Flash registered 2.51s, GPT-5 Nano logged 2.52s, and GPT-5 Mini reached 2.67s. The slowest models tested were the flagship Pro offerings: Gemini 3.1 Pro (preview) recorded a median speed of 3.70s, while Gemini 2.5 Pro reached 3.79s median latency.
Prices
Pricing for the measured Google Gemini models reflects clear stratification between entry-level Lite variants, mainstream Flash models, and flagship Pro tiers.
Gemini 2.5 Flash Lite represents the lowest cost endpoint measured, billed at $0.100 per million input tokens and $0.400 per million output tokens. Gemini 3.1 Flash Lite is priced at $0.250 per million input tokens and $1.50 per million output tokens, while Gemini 2.5 Flash costs $0.300 per million input tokens and $2.50 per million output tokens.
Moving to mid-tier offerings, Gemini 3.5 Flash costs $1.50 per million input tokens and $9.00 per million output tokens. Among high-capacity Pro models, Gemini 2.5 Pro is billed at $1.25 per million input tokens and $10.00 per million output tokens. Gemini 3.1 Pro (preview) carries the highest rate among reported pricing, set at $2.00 per million input tokens and $12.00 per million output tokens.
These measurements capture specific response times, uptime rates, billed prices, and benchmark accuracy under standardized test conditions. They do not prove long-term infrastructure stability, regional latency variations, or prompt performance beyond the tested evaluation parameters.
