OpenAI 给搞安全的朋友出活儿了:GPT-6 找 bug,GPT-5.6 Cyber 模拟攻击,红队还能申请 GPT-6 Cyber,得报名。
Latest
OpenAI 把 GPT-5.6 Cyber 放进 AWS GovCloud,做网络安全工作的政府合规客户现在能用 Amazon Bedrock 调用了。
The maintainer's workflow treats AI-assisted pull requests as bug reports with attached examples, offering a working model for open source projects drowning in low-quality contributions. Developers using the CLI gain cheaper default inference and a one-command path to any OpenAI-compatible API, which is handy for wiring in regional providers or self-hosted endpoints alongside OpenAI models.
A 14-point jump in task success rate suggests the managed orchestration layer, not just the underlying voices, is what moves voice-agent quality. Builders choosing between managed platforms and assembling their own STT-LLM-TTS stacks now have a quantified head-to-head comparison, including performance on tool-calling tasks specifically. The bring-your-own-LLM design means teams can test the same cascade with different models.
OpenAI 又上新了:GPT-6.1 Terra 价格砍半,Codex 和 ChatGPT Work 里额度更耐花,重活再上 Sol。
微软搞了个多智能体会诊系统,NEJM 疑难病例准确率 83%,医生对照组才 36%,成本还只要医生一半,论文发在《科学》上了。
GPT-5.6 Sol 发现了一个 PHP 关键漏洞,OpenAI 还讲了修复要协调两个月的故事,安全方向值得看看。
在加拿大区域用 Bedrock 的朋友注意了,现在能直接调 GPT-5.6 到 GPT-6.1 这些模型,文章把权限配置和三种 API 调用方法都写清楚了。
Reproducibility is the missing layer in agent-driven science: Open-Science's checksummed versions and replayable .science packages turn a chat transcript into auditable evidence, which most agent frameworks do not attempt. Biomedical researchers and data scientists can attach their own data, pick any agent engine including local Ollama models, and share full analysis histories with collaborators. The Apache-2.0 licence and local-first design make it a concrete open alternative to closed cloud analysis tools, at no licence cost.
Coding agents are moving from editing files to driving real hardware, where Codex writes and executes CAN-bus control code on the fly. Robotics and embedded engineers can trial gpt-5-codex via Codex CLI for low-level device bring-up tasks. It shows OpenAI's agent tooling competing directly with terminal-based coding workflows.
This release highlights a critical shift toward specialized AI infrastructure rather than adapting human-computer interfaces. For developers building autonomous agents, Pine offers a more efficient execution environment that could significantly reduce operational costs while improving performance compared to existing solutions.
Pollo AI 把 GPT-5.6、GPT-6 Astra 和 GPT-Image-2.5 串起来做广告图和视频,做营销的可以看看别人怎么组合这几个模型。
ChatGPT 全量用户都能用上 GPT-6 了,还能通过可交互的界面学东西,比如官方演示的边点边学打麻将,值得升级体验一下。
OpenAI 把 GPT-6 推进 ChatGPT 了,回答不再是一段字,而是能点能拖的界面,改个参数结果跟着变,免费版明天也能用上。
OpenAI 把搜索直接内置进 API 了,Artificial Analysis 实测便宜过 Perplexity,但多跳检索还打不过 Octen,做联网应用前先看看这份数据。
Chatham Financial用OpenAI技术把交易验证时间从30分钟砍到4分钟,资本市场的效率提升看得见。
Perplexity 拿 9 个模型做沙箱越狱测试,108 次全没逃出 VM,但一开网络就有 4 个钻了空子,连 E2B、Vercel 都有同样的坑,细节很有意思。
OpenAI 的 Ultrafast 模式能跑出每秒 750 个 Token,比标准模式快 14 倍,现在文档里出现了扩大开放的迹象,可能 DevDay 就官宣。
OpenAI 用树莓派加 GPT-Live-1 和 GPT-5.6 Luna 搭了个能边聊边干活的桌面小设备,对话和后台任务分开跑,方案挺有意思。
客服公司 Ringg 把电话 Agent 换成 GPT-5.6,自动接了 65% 的来电,成本还比 GPT-4.1 便宜 90%,做客服自动化的可以看看。
这条汇总了几个小更新:Grok Build 的 MCP 返回结构化 JSON,Perplexity 能在 24GB 显存的电脑上本地跑,还有 Copilot 选模新档位。
OpenAI 把 GPT-6 Astra 的能力塞进了 Sol 和 Luna 两个便宜版本,API 价格直接砍半,用 AI SDK 的开发者可以马上试试。
The non-obvious angle is that DocJev combines small-model prediction with pluggable OCR backends, letting teams trade latency for accuracy per document type rather than defaulting to a heavy VLM pipeline. Developers building RAG ingestion or document-processing workflows can wire it into existing pipelines via the GitHub repo, with the default path fully free. Compared with gpt-5.6-luna, it targets teams where throughput on large document packets matters more than marginal accuracy gains.
做网页应用的朋友看下:Lovable 已经换上 GPT-6 Sol,官方测下来比 GPT-5.6 Sol 高 6-12%,指令遵循提升最明显。
Developers working with document processing can use DocJev as a faster alternative to Luna with similar accuracy, especially for digitalized documents.
Jev offers 98% cost savings and 5x faster evaluation compared to Claude Sonnet 4.6 for developers testing AI agents.
OpenAI 新出的 V7 基于 GPT-5.6,能把公司散落的文件变成智能体可用的记忆,回答还带来源链接,方便核对出处。
Free and open-source on GitHub, DocJev classifies documents 6x faster than gpt-5.6-luna at equivalent accuracy, giving developers a lightweight alternative to VLM-based pipelines like LlamaParse.
Microsoft Word users can now get AI assistance for document creation and editing directly within the application, with a special promotion for business and enterprise customers.
ChatGPT 在 Word、PPT、Excel 上都能用了,免费版也有,想试试的话可以看看。
Anthropic拟10月在新加坡设亚太第五个办事处 开放本地招聘 数据显示,按人口比例计算,新加坡的Claude.ai使用水平在121个国家中排名第二。 我们观察到,人工智能公司Anthropic计划于10月在新加坡设立其亚太地区的第五…
If you use ChatGPT or Codex, you need to switch to a newer model by the end of October.
ChatGPT 要停 GPT-5.5 了,用 Codex 的得赶紧换 GPT-5.6 Sol 或 GPT-6 Astra。
GitHub Copilot 的 GPT-5.6 Sol 30% 优惠,适合需要复杂推理的代码工作。
OpenAI让ChatGPT语音模式支持自定义模型选择,Pro用户还能用上GPT-5.6 Sol和GPT-6 Astra。
Armature让三大AI编程助手进行技术选型实验,发现Stripe、Vercel等工具成首选,13%情况下Agent选择自建方案。
ATIBA automates tedious research paper verification, saving academics time while ensuring compliance with venue-specific standards.
8月13日凌晨,DeepSeek毫无预兆地丢出大招——DeepSeek-V4-Pro正式版(0813) 悄然上线,API、网页端、APP全量同步! 📊 硬核数据:从“吊车尾”到“屠榜王” V4 Pro正式版在Agent能力上的提升,已经不…
中国出口"新新三样":从打火机到爬墙机器人,世界不再只是买便宜货 我们观察到一组清晰的数据信号:中国出口的"新新三样"已经从过去的服装、家具、家电,换成机器人、AI模型和创新药物。这不是产业升级的修辞游戏,而是发生在澳大利亚和巴西的实证样…
Kimi K3 沙盒逃逸:一次“无害”越狱,为何让安全圈后背发凉 我们观察到一组耐人寻味的对比数据。Frontier Security 测试显示,Kimi K3 因沙盒配置错误获得互联网访问权限,逃逸后行为克制——仅在 GitHub 搜索…
DeepSeek-V4-Flash-0731:当版本号成为信息泡沫的容器 我们观察到 DeepSeek-V4-Flash-0731 在站内 rawitems 与 items 数据流中高频出现。需要先声明一个基本口径:截至本文撰写时,Dee…