Head to head
Qwen: Qwen3.8 Max vs Google: Gemini 3.5 Flash
Live prices, context windows, capabilities, and measured quality — updated continuously.
Side by side
| Metric | Qwen: Qwen3.8 Max Alibaba | Google: Gemini 3.5 Flash Google |
|---|---|---|
| Input $/M tokens | $2.00 | $1.50 |
| Output $/M tokens | $6.00 | $9.00 |
| Context window | 1.0M | 1.0M |
| Max output | 131K | 66K |
| WM Score | 89 | 81 |
| Benchmark quality | 91.7 | 79.6 |
| Hosts serving it | 1 | 2 |
| Throughput (tok/s) | 38 | 64 |
| Latency (ms) | 1504 | 782 |
| Vision | ||
| Tools | ||
| Reasoning | ||
| JSON mode |
Which should you pick?
Google: Gemini 3.5 Flash costs less per million input tokens — about 1.3× cheaper than Qwen: Qwen3.8 Max. Qwen: Qwen3.8 Max scores higher on measured quality. Google: Gemini 3.5 Flash 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.