GPT-6.2 Sol 编码接近 Astra,成本仅约五分之一
这次可以帮你省钱:日常写代码用 GPT-6.2 Sol 就够,在 FrontierCode 2.0 上只差 Astra 半分,成本却是五分之一。要是预算敏感的项目,直接用 Sol 起手,难题再切 Astra,比全用旗舰模型划算得多。
OpenAI 发布 GPT-6.2 系列,包含 Astra 和 Sol 两个模型,主打编码质量与效率的提升,官方称单位成本能获得更多智能。
在 Cognition 的 FrontierCode 2.0 基准上,Sol 开启最高推理档位后与 Astra 的差距不到半分,而每任务 API 成本约为 Astra 的五分之一。
OpenAI 建议日常编码任务优先使用 Sol,需要更高能力上限时再选 Astra。两个模型形成高低搭配的定价梯度,开发者可按任务难度灵活分配用量,降低总体 token 开销。
GPT-6.2 Astra and Sol continue to improve coding quality and efficiency, delivering more intelligence per dollar. For day-to-day coding, start with GPT-6.2 Sol. On Cognition’s FrontierCode 2.0 at max reasoning, it came within half a point of Astra at ~1/5 the API cost per task.📊 5 ⚡
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include GPT-6.2, OpenAI, coding assistants, reasoning models.
Primary source verification: published directly by @OpenAIDevs with first-party authority.