GPT-5.2-Codex vs Qwen3.6 Plus
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | GPT-5.2-Codex | Qwen3.6 Plus |
|---|---|---|
| Provider |  OpenAI |  Alibaba Qwen |
| score moyen | 70.6 | 70.9 |
| Prix d'entrée | $1.75 | $0.33 |
| Prix de sortie | $14.00 | $1.95 |
| Fenêtre de contexte | 400K tokens (~200 books) | 1.0M tokens (~500 books) |
| Sorti le | 2026-01-14 | 2026-04-02 |
| Code source ouvert | Proprietary | Open Source |
Scores de benchmark
8 benchmarks · GPT-5.2-Codex: 6, Qwen3.6 Plus: 2
| Benchmark | Catégorie | GPT-5.2-Codex | Qwen3.6 Plus |
|---|---|---|---|
| LiveBench — Agentic Coding | coding | 51.7 | 55.0 |
| LiveBench — Coding | coding | 83.6 | 78.2 |
| LiveBench — Data Analysis | reasoning | 78.2 | 69.9 |
| LiveBench — If | language | 66.5 | 58.3 |
| LiveBench — Language | language | 73.7 | 75.0 |
| LiveBench — Mathematics | math | 88.8 | 83.7 |
| LiveBench — Overall | knowledge | 74.3 | 70.8 |
| LiveBench — Reasoning | reasoning | 77.7 | 75.8 |