Gemini 2.5 Pro vs Qwen3 30B A3B Thinking 2507
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | Gemini 2.5 Pro | Qwen3 30B A3B Thinking 2507 |
|---|---|---|
| Provider |  Google DeepMind |  Alibaba Qwen |
| pontuação média | 56.2 | 66.2 |
| Preço de entrada | $1.25 | $0.08 |
| Preço de saída | $10.00 | $0.40 |
| Janela de contexto | 1.0M tokens (~524 books) | 131K tokens (~66 books) |
| Lançado em | 2025-06-17 | 2025-08-28 |
| Código aberto | Proprietary | Open Source |
Pontuações de benchmark
6 benchmarks · Gemini 2.5 Pro: 6, Qwen3 30B A3B Thinking 2507: 0
| Benchmark | Categoria | Gemini 2.5 Pro | Qwen3 30B A3B Thinking 2507 |
|---|---|---|---|
| OpenCompass — AIME2025 | math | 88.7 | 86.8 |
| OpenCompass — GPQA-Diamond | knowledge | 84.7 | 70.1 |
| OpenCompass — HLE | knowledge | 21.1 | 11.7 |
| OpenCompass — IFEval | language | 90.0 | 89.7 |
| OpenCompass — LiveCodeBenchV6 | coding | 71.3 | 59.4 |
| OpenCompass — MMLU-Pro | knowledge | 85.8 | 79.5 |