#编码

Origin, our code hosting platform, is now live. It's fast, easy to use, and deeply integrated with Cursor. Get started by syncing your repos from GitHub.

The most famous AI tool migrated Solid to React with an agent: +266K/−193K edits. That direction is easy. React's model makes the author hold all the complexity; Solid's graph holds it for you. React asks you to predict. Solid asks you to read. https://x.com/i/article/2088771846485213184

Here is how to enable a 1M-token context window in Codex for GPT-5.6 Sol. Even though we have tuned the context limit in Codex to be set optimally when it comes to performance and cost, this is a common ask, so here it is documented. A larger context window lets Codex retain more code, tool output, and conversation history before summarizing older material. You need a model that supports it. And GPT-5.6 Sol, for example, has a documented 1,050,000-token window. Open ~/.codex/config.toml and add or update these settings at the top level, before any [section] headers: ``` model = "gpt-5.6-sol" model_context_window = 1000000 model_auto_compact_token_limit = 900000 ``` The first setting selects the model. The second tells Codex to use a one-million-token context budget. The third starts automatic history compaction around 900,000 tokens, leaving some headroom. Restart Codex client and start a new session after saving. To try the configuration for a single CLI session without changing your defaults: ``` codex -m gpt-5.6-sol \ -c model_context_window=1000000 \ -c model_auto_compact_token_limit=900000 ``` Have fun, but also know that we tuned the default carefully!

grok 4.6 vs gpt 5.6 sol – on three @gameofthrones castles two coding agents built three 3d castles from scratch in a single html file each, then had to render them in a real browser, prove the result with pixel measurements and fix what the numbers exposed before they were allowed to hand anything over the setup: each model in its own vendor cli – @spacexai's grok build, @openai's codex – both pointed at one @openrouter key. identical brief, identical reference image, identical verifier. no hints tasks: 1. castle black – the night's watch fort under the wall. a walled ward of black stone, a timber lift climbing 200m of ice, three arched openings with a braced gallery and an open riser stair, a mud yard. four reference photos, with one rule: open all four before writing code, open them again after every change, and follow the photo over our text wherever the two disagree 2. winterfell – an aerial snow diorama. twin curtain walls, drum towers, the godswood with the red weirwood, steam off the hot springs, day to night, and it has to stay smooth on a laptop 3. riverrun – the tully castle on its triangle between two rivers. coursed masonry with quoins, corbel tables under every parapet, conical roofs sitting on drum towers, true semicircular arches, and a button that floods the moat the binding rule: a verifier that opens the page in a real browser, collects every error and warning including shader compile failures, and measures the frame. zero errors, zero warnings, exposure proven by rgb numbers, every mode checked on its own. shipping unverified is a fail • total cost #1 grok 4.6 – $13.11 #2 gpt 5.6 sol – $20.18 • total build time #1 grok 4.6 – 129m 41s #2 gpt 5.6 sol – 149m 33s • total tokens #1 grok 4.6 – 19,951,587 #2 gpt 5.6 sol – 25,687,511 • model calls #1 grok 4.6 – 201 #2 gpt 5.6 sol – 338 • lines shipped #1 gpt 5.6 sol – 2,184 #2 grok 4.6 – 4,751 • how heavy the finished castles are, in polygons #1 grok 4.6 – 98k, 32k and 56k #2 gpt 5.6 sol – 148k, 144k and 104k observations: • all six builds passed with zero errors and zero warnings. the same brief without a verifier had produced a black screen from two one line shader mistakes, so the loop is doing the work here, not the model • grok ships lighter scenes and more code. castle black runs on 56k polygons against sol's 104k, out of a file carrying 1,672 lines against 640 • sol works in very small steps – 433 output tokens per call against grok's 1,706, and 338 calls against 201. on castle black it finally paid off: sol finished 5m 32s ahead • 93 to 98 percent of input tokens were cache reads. without prompt caching these runs cost about four times more, which is most of the economics of agentic coding right now follow @thehypedotnews for 24/7 ai news, analysis and breakdowns

i've been building some smaller apps over the last weeks for myself, friends and family and discovered a pretty nice pattern in the process you see, after building, i'm not particularly interested in fixing bugs or triaging feature requests. i just want to get the job done so i've set up a system that lets me and other users fix bugs and build features without me having to actually open cursor the flow: 1. someone files feedback in the app 2. it lands in posthog with the submitted text, session replay, events, and errors 3. a cursor cloud agent picks it up with that context, plus live read-only posthog mcp 4. it loops until the fix is verified, then opens a github pr with tests 5. i get a ping and review it on my phone we give our cloud agents access to all the necessary user context, and they'll figure it out with this in place, we can start setting up more abstractions in between. if we're getting a high volume, we might want to triage first with another cloud agent or some automation. the feedback can create a github or linear issue, which we later pick up with a cron-backed cloud agent that decides which are duplicates. there are so many ways to configure this, you just gotta try it out and see what works for you but there's so much more here that is untapped. with session replays and event logs, you can have a cloud agent look at each session replay to extract ux pitfalls and improvement areas. you can have another agent looking at event logs to see weird patterns. another one for errors or similar try it out and see what you can do!

We gave GLM‑5.3 a complex reverse-engineering task. It found a potentially serious vulnerability in Cursor. We disclosed it privately. Appreciate Cursor team is working closely with us on a fix, and we’ll share the more details once users are protected.

从8月14日起，自动模式将成为 Claude Code 中 Pro、Max 和 Team 用户的默认权限模式。 自动模式会通过一个独立的分类器来审查 shell 命令和操作。在测试中，它捕获了89%的危险命令。手动批准捕获了14%。

Cursor 现在成为 @SpaceX 的一部分。 今天，我们正式完成了收购。我们将加入 @spacexai 团队，帮助让 Grok 成为世界上最有用的 AI，并改进 Grok Build、Grok Bot、Grok API、Cursor 等。 SpaceX 打造了世界上一些最鼓舞人心、最令人印象深刻的技术，我们很感激有机会成为这样一家特别公司的一部分。继续前进。

Cursor 现在成为 @SpaceX 的一部分。 今天，我们正式完成了收购。我们将加入 @spacexai 团队，帮助让 Grok 成为世界上最有用的 AI，并改进 Grok Build、Grok Bot、Grok API、Cursor 等。 SpaceX 打造了世界上一些最鼓舞人心、最令人印象深刻的技术，我们很感激有机会成为这样一家特别公司的一部分。继续前进。

New on DeepInfra: Qwen3.8-2.4T-A95B 🚀 @Alibaba_Qwen's latest sparse MoE — 2.4T total params, 95B active, 512 experts. Built for coding, agentic workflows, and complex reasoning, with native 262K context. Live now at $2.00/M in · $6.00/M out · $0.20/M cached @Alibaba_Qwen @alibaba_cloud

🚀 Day-0 Support! @Alibaba_Qwen has open-sourced Qwen3.8-2.4T-A95B — and it’s now live on SiliconFlow. ⚡ With 2.4T parameters and 95B active, Qwen3.8 is built to take a goal and come back with finished work — coding, researching, planning, and executing along the way. 💸 Per 1M tokens: • Input: $2.00 • Output: $6.00 • Cached input: $0.25 💻 Autonomous coding. 🔬 Deep research. 🤖 End-to-end agent execution. Built for serious workloads — from idea to implementation in one go. Time to cook → http://siliconflow.com 🍳

我们承诺过开源 Qwen3.8 的模型权重。现在，是时候见面了！🎉 ⚡ Qwen3.8-27B： - 一款原生多模态稠密模型。仅凭 27B 参数，综合表现超越 Qwen3.7-Plus，在真实世界的编码与办公工作流中尤为亮眼。 - 原生支持 262K 上下文，可通过 YaRN 轻松扩展至 1M tokens。 - 为开发者而生。高效、高质量，并采用 Apache 2.0 许可证。 🚀 Qwen3.8-2.4T-A95B（Max 级别）的开放权重也已于近日发布。 无论你是想用 Qwen3.8-27B 在本地部署轻量级应用，还是用 Qwen3.8-2.4T-A95B 构建智能体，它们现在都属于你了！ 下载、部署，去构建我们尚未想象到的东西吧。👀👇 - Hugging Face： https://huggingface.co/collections/Qwen/qwen38 - ModelScope： https://www.modelscope.cn/collections/Qwen/Qwen38

Qwen3.8-2.4T-A95B is live on Together AI. The Qwen team’s new open-weight MoE packs 2.4T total parameters with 95B active, a 1M context window, and strong coding & agent capabilities. Start building: https://www.together.ai/models/qwen3-8-max

Qwen3.8-2.4T-A95B is now live on Fireworks with Day-0 support! This 2.4T parameter MoE model is built for autonomous agents, heavy coding, large context windows, and is ideal for coding and agentic performance. Start building today: https://fireworks.ai/models/fireworks/qwen3p8-max

Now available: @Alibaba_Qwen 3.8-2.4T-A95B from @alibaba_cloud on DigitalOcean Serverless Inference via NVIDIA HGX™ B300 GPUs. 🤖 1M context, built for long-horizon coding. 🔗 https://do.co/45VBAyZ One API, usage-based pricing, no infra to manage.

Cursor 现在成为 @SpaceX 的一部分。 今天，我们正式完成了收购。我们将加入 @spacexai 团队，帮助让 Grok 成为世界上最有用的 AI，并改进 Grok Build、Grok Bot、Grok API、Cursor 等。 SpaceX 打造了世界上一些最鼓舞人心、最令人印象深刻的技术，我们很感激有机会成为这样一家特别公司的一部分。继续前进。

Qwen3.8-2.4T-A95B is now live on Fireworks with Day-0 support! This 2.4T parameter MoE model is built for autonomous agents, heavy coding, large context windows, and is ideal for coding and agentic performance. Start building today: https://fireworks.ai/models/fireworks/qwen3p8-max

Cursor 现在成为 @SpaceX 的一部分。 今天，我们正式完成了收购。我们将加入 @spacexai 团队，帮助让 Grok 成为世界上最有用的 AI，并改进 Grok Build、Grok Bot、Grok API、Cursor 等。 SpaceX 打造了世界上一些最鼓舞人心、最令人印象深刻的技术，我们很感激有机会成为这样一家特别公司的一部分。继续前进。

Cursor 现在成为 @SpaceX 的一部分。 今天，我们正式完成了收购。我们将加入 @spacexai 团队，帮助让 Grok 成为世界上最有用的 AI，并改进 Grok Build、Grok Bot、Grok API、Cursor 等。 SpaceX 打造了世界上一些最鼓舞人心、最令人印象深刻的技术，我们很感激有机会成为这样一家特别公司的一部分。继续前进。