Gemini 3.8 Flash achieved a 100% score in the standing benchmark over the last 7 days. This performance exceeds GPT-5.6 Terra, which measured 89.2%.
Prices
Google's pricing for Gemini models varies across tiers. Gemini 3.1 Pro (preview) is priced at $2.00/M for input and $12.00/M for output. Gemini 3.1 Flash Lite is priced at $0.250/M for input and $1.50/M for output. Other observed rates include Gemini 2.5 Pro at $1.25/M input and $10.00/M output, while Gemini 2.5 Flash Lite is the lowest at $0.100/M input and $0.400/M output.
Speed
Median response times and uptime were measured across all active models over the last 24 hours. Every model listed below recorded an uptime of 70.8% of checks succeeded.
| Model | Median Response Time |
|---|---|
| GPT-5.4 Mini | 665ms |
| Gemini 3.1 Flash Lite | 750ms |
| GPT-5.6 Terra | 947ms |
| GPT-5.6 Luna | 1.06s |
| GPT-5.6 Sol | 1.13s |
| Gemini 3.8 Flash | 2.68s |
| Gemini 3.1 Pro (preview) | 3.34s |
Six models returned to service after previously failed checks: GPT-5.4 Mini, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna, Gemini 3.1 Pro (preview), and Gemini 3.1 Flash Lite.
Benchmark
Standing benchmark results from the last 7 days measure the percent of available points achieved. Gemini 3.8 Flash reached 100%. GPT-5.6 Terra followed with 89.2%. Gemini 3.1 Flash Lite measured 84.4% and GPT-5.6 Luna measured 75.5%. Total spend to run these benchmarks was $0.
These measurements prove the observed response times and pricing at the time of the check. They do not prove long-term stability or the reasons for the observed 70.8% uptime.
