
Codex hit 4M active users, less than two weeks after hitting 3M. We will reset rate limits today!
enzyme · 独立开发
754 posts


Codex hit 4M active users, less than two weeks after hitting 3M. We will reset rate limits today!

我刚把Gpt认证了下,才能用新的画图API。对比来了,同样是岳飞《满江红》草书作品,左边是banana-2,右边是gpt-image-2,后者是碾压啊!

What goes into the system prompt vs. what goes into an Agent Skill? Agent’s system prompt is for identity, constraints, and persistent context. • What the agent is and what it should always or never do • How it should generally approach problems (reasoning style) • Persistent context needed for every interaction Agent Skills are for capabilities the agent can choose to invoke when relevant. • Dynamic context • Performing actions with side effective (e.g., calling APIs) • Anything that’s optional depending on the situation Practical rules of thumb: If it’s a specific action the agent does, it goes in a skill. If it’s the identity of who the agent is and how it thinks, it goes in the system prompt. If it changes per-request, it goes in a skill. If it's stable across all requests, it goes in the system prompt.

Exciting news - GPT-Image-2 by @OpenAI has claimed the #1 spot across all Image Arena leaderboards! A clean sweep with a record-breaking +242 point lead in Text-to-Image - the largest gap we’ve seen to date. - #1 Text-to-Image (1512), +242 over #2 (Nano-banana-2 with web-search aka gemini-3.1-flash-image) - #1 Single-Image Edit (1513), +125 over #2 (Nano-banana-pro aka gemini-3-pro-image) - #1 Multi-Image Edit (1464), +90 over #2 (Nano-banana-2) No model has dominated Image Arena with margins this wide. Huge congratulations to @OpenAI on this major breakthrough in image generation! More performance breakdowns by category in the thread below.

OpenAI 用 白嫖党养飞轮,Anthropic 连 Opus 付费用户都供不起了 OpenAI 和 Anthropic 这一仗,根本就是 Windows 打 Mac 的翻版。 你看最近多离谱——Codex 注册机满天飞,官方不但不管,Sam Altman 还"喜迎" 400 万新用户,顺手把额度全重置了。 这操作你眼熟吗? 当年微软对 Windows 盗版什么态度?默许,甚至纵容。 不是心大,是因为比尔·盖茨比谁都清楚:装机量才是最大的护城河。 今天 OpenAI 的阳谋: 你可以白嫖,可以钻空子,但只要你在用,你的数据就在帮我进化。 但这里有个绝大多数人忽略的点——这游戏不是谁都能玩的。 放任上百万人白嫖、重置额度、甚至默许注册机泛滥,背后是要真金白银烧算力的。Anthropic 敢这么跟吗?够呛。Opus 的降智风波 已经说明了一切——连自己付费用户的算力都供不起,哪还有余力接 400 万白嫖党? 用户越多 → 数据越多 → 模型越强 → 工具越好用 → 用户越多。 等这个飞轮转起来,正版和盗版的边界还重要吗? 底子够厚,才有资格不计较眼前输赢。它要的是整个市场,不是你那点订阅费。




Codex hit 4M active users, less than two weeks after hitting 3M. We will reset rate limits today!


Meet Kimi K2.6: Advancing Open-Source Coding 🔹Open-source SOTA on HLE w/ tools (54.0), SWE-Bench Pro (58.6), SWE-bench Multilingual (76.7), BrowseComp (83.2), Toolathlon (50.0), Charxiv w/ python(86.7), Math Vision w/ python (93.2) What's new: 🔹Long-horizon coding - 4,000+ tool calls, over 12 hours of continuous execution, with generalization across languages (Rust, Go, Python) and tasks (frontend, devops, perf optimization). 🔹Motion-rich frontend - Videos in hero sections, WebGL shaders, GSAP + Framer Motion, Three.js 3D. 🔹Agent Swarms, elevated - 300 parallel sub-agents × 4,000 steps per run (up from K2.5's 100 / 1,500). One prompt, 100+ files. 🔹Proactive Agents - K2.6 model powers OpenClaw, Hermes Agent, etc for 24/7 autonomous ops. 🔹Claw Groups (research preview) - bring your own agents, command your friends', bots & humans in the loop. - K2.6 is now live on kimi.com in chat mode and agent mode. For production-grade coding, pair K2.6 with Kimi Code: kimi.com/code - 🔗 API: platform.moonshot.ai 🔗 Tech blog: kimi.com/blog/kimi-k2-6 🔗 Weights & code: huggingface.co/moonshotai/Kim…