Gemini 2.5 Pro vs Qwen3 Next 80B A3B Instruct
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | Gemini 2.5 Pro | Qwen3 Next 80B A3B Instruct |
|---|---|---|
| Provider |  Google DeepMind |  Alibaba Qwen |
| score moyen | 56.2 | 54.4 |
| Prix d'entrée | $1.25 | $0.09 |
| Prix de sortie | $10.00 | $1.10 |
| Fenêtre de contexte | 1.0M tokens (~524 books) | 262K tokens (~131 books) |
| Sorti le | 2025-06-17 | 2025-09-11 |
| Code source ouvert | Proprietary | Open Source |
Scores de benchmark
10 benchmarks · Gemini 2.5 Pro: 10, Qwen3 Next 80B A3B Instruct: 0
| Benchmark | Catégorie | Gemini 2.5 Pro | Qwen3 Next 80B A3B Instruct |
|---|---|---|---|
| Artificial Analysis — Agentic Index | speed | 32.7 | 14.2 |
| Artificial Analysis — Coding Index | speed | 31.9 | 15.3 |
| Artificial Analysis — Quality Index | speed | 34.6 | 20.1 |
| Chatbot Arena Elo — Overall | arena | 1448.2 | 1401.6 |
| OpenCompass — AIME2025 | math | 88.7 | 69.2 |
| OpenCompass — GPQA-Diamond | knowledge | 84.7 | 74.1 |
| OpenCompass — HLE | knowledge | 21.1 | 8.0 |
| OpenCompass — IFEval | language | 90.0 | 87.6 |
| OpenCompass — LiveCodeBenchV6 | coding | 71.3 | 54.8 |
| OpenCompass — MMLU-Pro | knowledge | 85.8 | 81.3 |