2026-10-28 AI 日报

Open-source Grok Bot alternative with a virtual machine that bots can use

Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable decision ledger for every agent step.…

DeepSeek Harness: Everything is a Plugin.

A local-first desktop workspace for persistent AI teammates. Run Codex, Claude, and Grok with dedicated workspaces, task queues, file sharing, browser…

Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.

DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

Macro is a unified workspace for teams: email, chat, docs, tasks, agents, calls, and CRM — @-linked together with shared AI memory.

A collection of AI agent skills for working with Expo projects and Expo Application Services

Consider it done. The open-source AI agent that works out of the box · 想到，就能做到。开源、开箱即用的 AI Agent。

The design vocabulary for your coding agent. Steer with polish and bolder, review with critique and audit, catch AI slop before it ships.

Interactive world generation requires models to follow successive commands while preserving relevant history and action consequences. We present Hy-Nomas, a…

Multimodal intelligence can be viewed as human-guided co-evolution between worlds and agents through interaction. However, scaling this process faces two…

Memory is increasingly used in long-horizon agents, yet the net value of building memory over direct retrieval from raw history remains largely unclear.…

Training a spatial VLM to answer which side an object lies on does not require it to preserve the metric displacement that determines that answer, while…

Discrete visual tokens represent vision in a discrete form analogous to language, offering a native interface for multimodal modeling under a shared…

Recent test-time training (TTT) frameworks have enabled open-weight large language models (LLMs) to outperform inference-only frameworks powered by frontier…

Long-horizon LLM agents interact with their environments through complete assistant turns while generating each action token by token. We observe that critic…

In long-term interactions with large language model (LLM) agents, users reveal habits through their choices and actions. These habits capture implicit user…

Diffusion video models produce striking clips, but they cannot reliably render an exact formula, a correct chart, a rule-consistent game or narration…

Vision--Language--Action (VLA) models generate robot actions from visual observations and language instructions, but their understanding of physical dynamics…

Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities. Today, we’re announcing the next frontier model:…

A transcript can capture every spoken word and still leave users or applications wondering who said what, how to translate it, or how to format…

We introduce a sleep phase in continual learning, enabling large language models to continuously integrate new facts and enhance reasoning while mitigating…

I’ve written before about how November 2025 was an inflection point for coding agents—the point where Opus 4.5 and GPT-5.1 crossed a tipping point that meant…

BREAKING: Claude Code found 14,219 photos of my face on my Android that I never took. Phone was secretly taking them almost every time I picked it up. Even…

I noticed something interesting in the DeepSeek Harness repository 👀 In provider.spec.ts, there are references to deepseek-v5, deepseek-v5-open, and…

So...what would you like to see in our future Gemma models?

From the reported benchmarks, it looks to be somewhat better than Opus 5.5. Funny conspiracy theory: that Google delayed its release until Artificial…

lot of people making stuff up so here's some more context you can imagine there's hidden enterprise demand for open source models that isn't visible but if…

Model & weights: BrunoHays/embeddinggemma-2-nullspace-audio I wanted to share an experiment on google/embeddinggemma-2 (577M) for MAEB, the 30-task audio…

This might have implications for tool calling and computer use. Write-up: https://divagr.com/blog/qwenputer HF:…

Fun fact: if Gemini 4 Carbon would have been released yesterday, it would not have been #1 on Artificial Analysis. On AA v4.3.2, it scores 57 against Opus 5.5…

https://x.com/osanseviero/status/2115351234811249111

CodeRabbit uses GPT-6.1 Sol to investigate security risks across complex codebases. Its security agent helps developers review findings and proposed fixes.

This post walks through the Agents CLI evaluation loop for systematically improving agent performance through continuous testing and grading. It demonstrates…

Today, we are rolling out immersive explorations in the new Gemini student hub.

Genie 4: our new world model unlocking unprecedented realism in real-time simulation.

OpenAI shares model results on the Kakeya conjecture, Exponential Time Hypothesis, and quantum chromodynamics—and how researchers can verify and build on them.

Z.ai GLM 5.3 Flash (zai-glm-5-3-flash) is now available. A multimodal 320B-parameter (18B active) Mixture-of-Experts model from Z.ai with a 1M-token context…

You can now set content guidelines on text API requests. Add a guidelines array to a Chat Completions or Responses API request to set how Grok handles content…

Meta is bringing Superintelligence models, research funding, and direct researcher collaboration to US universities that are using our models to solve…