@Angaisb_ Sometimes when I'm using 5.5 thinking,it says it's 5.4 mini
English
Vancouver Liu
12 posts
GPT-5.5自己给出的benchmark分数那么高，这被第三方一测咋这么尴尬
Arena.ai@arena
GPT-5.5 by @OpenAI is now live in the Arena, landing across multiple leaderboards. Here’s how it ranks by modality: - Code Arena (agentic web dev): #9, a strong +50pt jump over GPT-5.4 - Document Arena (analysis & long-content reasoning): #6, on par with Sonnet 4.6 - Text Arena: #7, Math #3, Instruction Following: #8 - Expert Arena: #5 - Search Arena: #2 - Vision Arena: #5 Strong, well-rounded performance, especially in Code (+50 pts vs GPT-5.4). Congrats to @OpenAI on the release. Full category breakdowns by modality in the thread.
中文
English
GPT-5.5 by @OpenAI is now live in the Arena, landing across multiple leaderboards.
Here’s how it ranks by modality:
- Code Arena (agentic web dev): #9, a strong +50pt jump over GPT-5.4
- Document Arena (analysis & long-content reasoning): #6, on par with Sonnet 4.6
- Text Arena: #7, Math #3, Instruction Following: #8
- Expert Arena: #5
- Search Arena: #2
- Vision Arena: #5
Strong, well-rounded performance, especially in Code (+50 pts vs GPT-5.4).
Congrats to @OpenAI on the release. Full category breakdowns by modality in the thread.
OpenAI@OpenAI
Introducing GPT-5.5 A new class of intelligence for real work and powering agents, built to understand complex goals, use tools, check its work, and carry more tasks through to completion. It marks a new way of getting computer work done. Now available in ChatGPT and Codex.
English
GPT-5、Claude 4.5、Gemini 3，我用 30 个真实任务测出了这个结论
三家旗舰模型我都付了钱，过去一个月跑了 30 个真实工作任务，给你一个不带粉丝滤镜的横评。
长文写作 / 报告生成：Claude 4.5 ＞ GPT-5 ＞ Gemini 3 Claude 的文风最像人，逻辑链条最稳，写 5000 字以上的方案不会自相矛盾。GPT 偏"模板感"，Gemini 经常跑题。
写代码 / Debug：Claude 4.5 ≈ GPT-5 ＞＞ Gemini 3 两家几乎打平，但风格不同：Claude 一次性写完整的方案，GPT 喜欢小步迭代。Gemini 在复杂项目里还是会犯低级错误。
做研究 / 跨网搜索：Gemini 3 ＞ GPT-5 ＞ Claude 4.5 Gemini 接 Google 全家桶，找资料速度断层第一。Claude 没有原生联网在这个场景吃亏。
多模态（图、视频、PDF）：GPT-5 ＞ Gemini 3 ＞ Claude 4.5 GPT 的视觉理解最强，做图表分析、扫 PDF 都很稳。
性价比：Claude（API 调用最便宜）＞ Gemini（免费额度最大）＞ GPT（订阅最贵）
最终建议：
· 一个人只买一个 → Claude
· 买两个 → Claude + Gemini（免费）
· 全都要 → 你应该是创业者，三个都买就对了
工具不是越多越好，是用得越深越好。
中文
Vancouver Liu retweetledi
我原以为，Claude 做事是要脸的
现在看来，天下乌鸦一般黑
做产品的粉丝，不做品牌的粉丝
——谨记于心
“过去一个月，一些用户报告 Claude Code 的质量有所下降。我们进行了调查，并发布了一篇关于我们发现的三个问题的事后分析。
所有问题已在 v2.1.116+ 版本中修复，我们已为所有订阅者重置了使用限制。”
ClaudeDevs@ClaudeDevs
Over the past month, some of you reported Claude Code's quality had slipped. We investigated, and published a post-mortem on the three issues we found. All are fixed in v2.1.116+ and we’ve reset usage limits for all subscribers.
中文
Across languages, Grok still leads where it matters: real usage.
🌍 Grok Code Fast 1 ranks #1 on OpenRouter
📊 136B tokens processed in natural language
🥇 Top share among individual models
This isn’t about one task or one region.
It’s about global, day-to-day adoption across languages.
When users choose freely, Grok stays on top.
English
