AstroHan

Exciting news: Gemini 4 Carbon (High) from @GoogleDeepMind just made a huge debut on Arena, landing #1 in Text Arena, #2 in Agent Arena, and #4 in Code Arena: WebDev. Back at the top, Gemini 4 Carbon delivers frontier intelligence and reshapes the Pareto frontier across three arenas. In Agent Arena, Gemini 4 Carbon ranks #2 with a +16.2% net improvement across +21.9K real-world agentic sessions from our global community, surpassing top models GPT-6 Astra and Fable-5.1. At $1.80 median cost per task, Gemini 4 Carbon (High) costs 72% less than the #1 ranked Claude Opus 5.5 (Max). Its standouts by key signals: - #1 Steerability (+18.6%): can the model course-correct when you push back? - #2 Praise vs Complaint (+37.9%): implicit sentiment in users' reactions - #3 Confirmed Success (+18.7%): explicit “yes, that worked” from users In Code Arena: WebDev, Carbon scored 1768 pts landing #4: - Just 2 pts below #3 Claude Sonnet 5.5 (xHigh), at 1770 pts - 89 pts above #11 Gemini 4 Argon (High), at 1679 pts By category, it ranks #3 in Gaming and Simulations, and #5 in Brand & Marketing. In Text Arena, Gemini 4 Carbon (High) is the new #1 with 1528 pts, now ahead of Gemini 4 Argon (High) at #2 with 1520 pts. Congrats to the @GoogleDeepMind team on this huge milestone!

现在几乎主要的 Agent 都是用 Electron 方案，Codex、Claude...一旦同时跑的 Session 和工作多了，性能消耗和资源占用都很大，尤其是 Web 渲染的各种动画。 我们在把整个 Maka 重写为 Rust + GPUI。为未来几十/上百/上千个 agent 并行做好准备。 已经几乎可用了： github.com/apache/maka/tree/…

Sam Altman: OpenAI's advantage is their internal models are way ahead of public models. I asked how far ahead they are:

如果maka tui是这样的，是不是很酷 demo使用 nsh.dev 编写，在我的fork添加了tui backend。在支持kitty sixel iterm2 图形协议的终端上可以显示成这样，不支持的终端回退到 unicode 字符。原理也比较简单，就是利用nsh的gpui后端生成可提取的像素，然后注入到nsh tui后端即可

👋 Hi everyone! Long time no see We're the LlamaFactory team. After passing 70k stars on GitHub, we're now building the "LlamaFactory" of recursive self-improvement (RSI), towards a unified, efficient, stable, and flexible framework for agent self-evolution. Excited to share our team's latest work with you all. TL;DR: 🐧 We've open-sourced PenguinHarness, a multi-agent self-evolution framework that brings every RSI algorithm under one roof, over 2,600 GitHub stars. 📖 To get a fuller picture of existing RSI work, we went through 140 papers, organized and categorized them into the awesome-rsi repo, and published a survey paper. 📊 To compare self-evolution efficiency across algorithms fairly, we built GDPevo, a self-evolution benchmark covering 6 enterprise scenarios and 240 tasks, and used it to evaluate 6 mainstream agents. The details: Following the same philosophy as LlamaFactory, we want to bring all RSI-related data, models, algorithms, and benchmarks into a unified open-source harness, so everyone can try and debug RSI more easily. We've already integrated 5 RSI algorithms and 7 benchmarks, and we're aiming for 100+ down the road. Borrowing from DSH's "everything is a plugin" approach, we've packaged every algorithm and benchmark as a plugin. They're fully decoupled, so users can freely mix and match any RSI algorithm to evolve different agents. We also built a modern desktop client that installs in one click on any OS and runs locally, so you can grab the latest version from our website. Website: penguin.ooo/ OSS Repo: github.com/Prism-Shadow/peng… To better understand where the academic work stands today, we open-sourced the awesome-rsi repo and published a survey paper covering 140 recent works. We also mapped the trajectory of RSI against that of deep learning — many concepts line up one-to-one, only now the subject is an agent. OSS Repo: github.com/Prism-Shadow/awes… Paper: arxiv.org/abs/2610.14625 To better evaluate how different harnesses and methods perform at self-evolution, we built the GDPevo benchmark, spanning 6 real-world enterprise scenarios and 240 tasks. With a train/val split, it fairly measures an agent's self-evolution ability while guarding against overfitting. OSS Repo: github.com/Prism-Shadow/GDPe… Paper: arxiv.org/abs/2608.03764 This post isn't just about our work. We're also hoping to invite friends from academia and industry to join our open-source project. If you'd like to see your name on an open-source project and co-author papers with us, DM us and let's talk!

A few big upgrades for Chrome users today! We’re bringing the power of computer use to anyone who uses Chrome. First up: ChatGPT can now use your tabs in Chrome to make answers more useful and get things done for you. If you’re in Chat, it can look across your open tabs and gather info. If you’re in Work, ChatGPT can control your Chrome tabs to get things, like sending messages, filling out forms, or downloading files. It can even help you organize and group that pile of tabs you have open.

I've actually stopped using the word "harness." I now think in terms of two things: agent runtimes and agent applications. Agent runtimes should be very small. Their core job is to connect an LLM API to an execution environment. That environment could be a full OS or a language VM, like a JavaScript runtime. Almost all that matters here is robustness. An agent runtime must be able to keep an agent running forever. Literally forever. Agent applications are everything else. Yes, literally every piece of software in the world. If software isn't built for agents, they use it through computer use or browser use. These capabilities can be built into the agent runtime or provided by another agent application. If software is agent-native, agents should be able to use it directly. The experience of an agent using that software—what we call "agent experience"—needs to be carefully designed, at least for now, to make efficient use of both time and tokens. Another key piece is agent identity. Agent applications should recognize each agent individually, rather than just the browser or computer it's using. This allows each agent's actions to be recorded in detail and audited in real time or after the fact. Everything we're building at @botiverse follows from this view.

发现个很有意思的事情，在中文 X 中，Claude Opus 5.5 几乎是碾压 GPT 6.1 Sol，所有人都说吊打。 但在英文的 X 上，现在 Claude Opus 5.5 对比 GPT 6.1 Sol 大概是 6/4 开或者 7/3 开，其实 GPT 6.1 Sol 的口碑也是不错的。 整体上大家还是认可 Opus 5.5 是目前最好的模型，但没有这么一边倒。