Primary sourceLxfater
View sourceEvent ·Multi-source
GPT-6.2 Sol 编程跑分较上一代下降 3 分,API 成本降约 19%
Artificial Analysis 用 Codex 对 GPT-6.2 Sol 和上一代 Sol 进行了对比测试,两者都开到 max 设置。结果显示 GPT-6.2 Sol 的编程跑分为 49,比上一代的 52 低了 3 分,降幅约 5.8%。成本方面,每题 API 价格从 $1.51 降到 $1.22,便宜了约 19%。这组数据勾勒出一个典型的取舍:OpenAI 用一部分编程能力换取了更低的调用成本。对于以 Codex 跑分作为选型参考的开发者来说,是否升级取决于对价格和编程质量的权重分配。
Current read
GPT-6.2 Sol 在 Codex 测试里比上一代少 3 分,但每题成本从 $1.51 降到 $1.22。写代码为主的项目可能不适合升级,跑量大、对编程精度要求不高的场景反而能省钱。选型前建议先看这组对比数据。
11 sources61° AI heatLast updated 2026/11/18 02:24:21
Evidence
11 sourcesRelated source 1Artificial Analysis
View sourceRelated source 2宝玉
View sourceRelated source 3Decoder
View sourceRelated source 4TestingCatalog
View sourceRelated source 5The Rundown AI
View sourceRelated source 6kimmonismus
View sourceRelated source 7marktechpost
View sourceRelated source 8IT之家
View sourceRelated source 9OpenRouter
View sourceRelated source 10@koltregaskes
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.