Head to head
OpenAI: GPT-5.4 vs Google: Gemini 3.7 Flash
Live prices, context windows, capabilities, and measured quality — updated continuously.
Side by side
| Metric | OpenAI: GPT-5.4 Openai | Google: Gemini 3.7 Flash Google |
|---|---|---|
| Input $/M tokens | $2.50 | $0.750 |
| Output $/M tokens | $15.00 | $3.75 |
| Context window | 1.1M | 1.0M |
| Max output | 128K | 66K |
| WM Score | 85 | 82 |
| Benchmark quality | 84 | 79 |
| Hosts serving it | 3 | 2 |
| Throughput (tok/s) | 58 | 166.5 |
| Latency (ms) | 2976 | 1103 |
| Vision | ||
| Tools | ||
| Reasoning | ||
| JSON mode |
Which should you pick?
Google: Gemini 3.7 Flash costs less per million input tokens — about 3.3× cheaper than OpenAI: GPT-5.4. OpenAI: GPT-5.4 scores higher on measured quality. OpenAI: GPT-5.4 takes a larger context window, which matters for long documents or big codebases. Prices here are live list prices; the same model can cost less on a different host, so check the model page for the cheapest offering.