ProductsMulti-source74°

GPT-6.2 Astra 与 GPT-6.2 Sol 在 GitHub Copilot 正式可用

Why we picked it

OpenAI 给 GitHub Copilot 上新了两款编码模型,重点强化代码评审:Sol 找真实问题的数量接近上一代两倍,Astra 干活更省 token。写代码依赖 Copilot 评审的朋友可以直接在 CLI 和 VS Code 里切换试试。

OpenAI 的两款新模型 GPT-6.2 Astra 和 GPT-6.2 Sol 已在 GitHub Copilot 正式开放(GA),定位为智能体编程和代码评审场景,官方称两者的编码判断力强于此前的 GPT-6 系列。用户可在 GitHub Copilot 应用、CLI 和 VS Code 中直接选用这两款模型。

Image · GitHub
Body

具体提升方面:GPT-6.2 Astra 在早期测试中比 GPT-6 Astra 发现了明显更多的代码评审问题,同时完成智能体编程任务消耗的 token 更少;GPT-6.2 Sol 在代码评审中捕获的真实问题数量接近 GPT-6.1 Sol 的两倍,且保持了发现的准确性。这表明 OpenAI 在针对代码评审的精确率和智能体任务的 token 效率上做了专项优化。

此前 GitHub Copilot 的模型阵营已覆盖 Anthropic、Google 等多家供应商,本次更新延续了模型竞技场的竞争态势。对重度使用 Copilot 做代码评审和智能体编程的开发者来说,新模型可能在同样预算下覆盖更多代码变更审查。

Original · GitHub

GPT-6.2 Astra and GPT-6.2 Sol from @OpenAIDevs are now generally available in GitHub Copilot. Designed for agentic coding and code review, both models demonstrate stronger coding judgment than earlier GPT-6 models. 🌌 GPT-6.2 Astra: In early testing, it caught significantly more issues in code review than GPT-6 Astra and completed agentic coding tasks using noticeably fewer tokens. ☀️ GPT-6.2 Sol: In early testing, it caught nearly twice as many real issues in code review as GPT-6.1 Sol while keeping its findings accurate. Try both in the GitHub Copilot app, CLI, and @code . github.blog/changelog/2026… R📊 6 ⚡

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include GPT-6.2, GitHub Copilot, OpenAI, coding assistants.

Source attribution: curated from GitHub (@github). Cross-source corroboration: 11 independent sources have covered this development (including TestingCatalog, Genspark, lmarena.ai), confirming industry-wide signal strength.