Gemini 2.5 Pro vs Qwen3 32B
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | Gemini 2.5 Pro | Qwen3 32B |
|---|---|---|
| Provider |  Google DeepMind |  Alibaba Qwen |
| score moyen | 56.2 | 58.2 |
| Prix d'entrée | $1.25 | $0.08 |
| Prix de sortie | $10.00 | $0.24 |
| Fenêtre de contexte | 1.0M tokens (~524 books) | 41K tokens (~20 books) |
| Sorti le | 2025-06-17 | 2025-04-28 |
| Code source ouvert | Proprietary | Open Source |
Scores de benchmark
8 benchmarks · Gemini 2.5 Pro: 8, Qwen3 32B: 0
| Benchmark | Catégorie | Gemini 2.5 Pro | Qwen3 32B |
|---|---|---|---|
| Aider polyglot | coding | 83.1 | 40.0 |
| Chatbot Arena Elo — Overall | arena | 1448.2 | 1347.0 |
| OpenCompass — AIME2025 | math | 88.7 | 70.3 |
| OpenCompass — GPQA-Diamond | knowledge | 84.7 | 67.3 |
| OpenCompass — HLE | knowledge | 21.1 | 8.5 |
| OpenCompass — IFEval | language | 90.0 | 86.0 |
| OpenCompass — LiveCodeBenchV6 | coding | 71.3 | 57.6 |
| OpenCompass — MMLU-Pro | knowledge | 85.8 | 78.0 |