开发者分享半年使用体验：Claude Code + Opus 5.5 成为日常主力，Codex 在计算机操作上仍占优
这是一线开发者的真实对比：Claude Code 加 Opus 5.5 写代码更强，Codex 则在 computer use 上更稳，两个新玩具 ChatGPT Dot 和 Claude Duet 也各有看点。正在纠结选哪套编程工具的人可以参考。
一位开发者在 X 上分享了他 2025 年的编程助手使用情况：过去一年在 Codex/GPT 与 Claude Code/Opus 之间来回切换，近两个月主要使用 Claude Code 搭配 Opus 5.5 作为日常主力工具。
他认为 Opus 5.5 是目前效率和智能表现最好的模型，GPT 虽然不错但仍有差距。不过 Claude 的短板在于 computer use 能力，这方面 GPT 目前无人能敌，他也希望 Anthropic 能补齐这一环。
在新产品方面，他提到欧洲目前能用到 ChatGPT Dot，而他个人更感兴趣的是同样尚未在欧洲上线的 Claude Duet。他仍期待 GPT-6.2 的表现，但现阶段他的选择是 Claude。
For me it has been a back and forth between Codex/GPT and Claude Code/ Opus this year.
For two months now I mainly use Claude Code and Opus 5.5. Its been my daily driver. Its the most efficient and intelligent model to date and GPT is good, but not close to Opus 5.5 imho.
What's holding back Claude right now is computer use. GPT is unbeaten here. And id love to see Anthropic catching up here.
Right now it looks like Europe gets access to ChatGPT Dot. But honestly, im more interested in Claude Duet right now (which ofc is also not available in Europe rn)
Anyways, still hopes for GPT-6.2. But right now my winner is Claude.
This report is categorized under AITOP's "How-to" 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. Key entities and technical topics under continuous tracking include Claude Code, Opus 5.5, Codex, GPT.
Source attribution: curated from kimmonismus (@kimmonismus).
