Six OpenAI models returned to operational status in our 24-hour monitoring window after previously failing health checks. GPT-5 Nano, GPT-5 Mini, GPT-5.4 Mini, GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.6 Luna all stabilized to achieve 100% uptime across hourly checks. Simultaneously, Gemini 3.6 Flash and GPT-5 Mini tied for the top position on the 7-day standing benchmark, securing 100% of available points at $0 in total spend.
Prices
Pricing across Google's Gemini lineup spans from budget Flash Lite models up to preview Pro releases. [Gemini 2.5 Flash Lite](/models/google-gemini-2-5-flash-lite) remains the lowest-cost option measured, priced at $0.100 per million input tokens and $0.400 per million output tokens. Gemini 3.1 Flash Lite offers higher entry capabilities at $0.250 per million input tokens and $1.50 per million output tokens.
For mid-tier deployment, Gemini 2.5 Flash costs $0.300 per million input tokens and $2.50 per million output tokens, whereas Gemini 3.5 Flash commands $1.50 per million input tokens and $9.00 per million output tokens. In the heavy-workload tier, Gemini 2.5 Pro is billed at $1.25 per million input tokens and $10.00 per million output tokens. The premium Gemini 3.1 Pro (preview) sits at the top of the observed price scale at $2.00 per million input tokens and $12.00 per million output tokens.
| Model | Input Cost ($/M) | Output Cost ($/M) |
|---|---|---|
| Gemini 2.5 Flash Lite | $0.100 | $0.400 |
| Gemini 3.1 Flash Lite | $0.250 | $1.50 |
| Gemini 2.5 Flash | $0.300 | $2.50 |
| Gemini 2.5 Pro | $1.25 | $10.00 |
| Gemini 3.5 Flash | $1.50 | $9.00 |
| Gemini 3.1 Pro (preview) | $2.00 | $12.00 |
Speed
Latency measurements over the last 24 hours showed every tracked model maintaining 100% uptime across hourly tests. Gemini 2.5 Flash Lite recorded the fastest median response time overall at 528ms. GPT-5.4 Mini followed closely as the fastest non-Google model at 661ms median latency.
Mid-speed performance was led by Gemini 2.5 Flash at 909ms and Gemini 3.1 Flash Lite at 913ms median response times. The GPT-5.6 family showed tight clustering: GPT-5.6 Luna registered 1.03s, GPT-5.6 Terra logged 1.09s, and GPT-5.6 Sol recorded 1.20s. Higher-capacity models showed slower response times, with Gemini 3.6 Flash at 1.81s, GPT-5 Mini at 2.09s, Gemini 3.5 Flash at 2.21s, and GPT-5 Nano at 2.38s. Gemini 3.1 Pro (preview) and Gemini 2.5 Pro registered the longest median latencies at 3.59s and 3.85s, respectively.
Benchmark
Over the last 7 days, benchmark results demonstrated distinct performance tiers across models evaluated with $0 in recorded spend. Both Gemini 3.6 Flash and GPT-5 Mini achieved perfect scores, taking 100% of available points. GPT-5.6 Terra earned the next highest standing at 90.6%, followed by Gemini 3.1 Flash Lite at 87.5%. Further down the standing benchmark, GPT-5.6 Luna reached 71.9% of available points, while Gemini 2.5 Flash recorded 67.7%.
| Model | Benchmark Score (%) | Total Spend ($) |
|---|---|---|
| Gemini 3.6 Flash | 100% | $0 |
| GPT-5 Mini | 100% | $0 |
| GPT-5.6 Terra | 90.6% | $0 |
| Gemini 3.1 Flash Lite | 87.5% | $0 |
| GPT-5.6 Luna | 71.9% | $0 |
| Gemini 2.5 Flash | 67.7% | $0 |
These quantitative readings reflect direct performance across automated hourly checks and standard test suites during the specified period. They confirm current latency, price schedules, and accuracy ratings under test conditions, but do not predict long-term reliability or behavior under unmeasured production environments.
