2026-10-23 AI 日报

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。

面向 AI 创作的开源无限画布工作台，集成 AI 生图、参考图编辑、视频生成、Agent 智能助手、画布编排、对话创作、提示词库与素材管理等能力，支持可视化创作流程与多 Agent 协同工作。兼容 OpenAI 接口生态，支持 chatgpt2api、grok2api、flow2api、newapi 等渠道接入。

claude-red is a curated library of offensive security skills designed for the Claude skills system. Each skill is a structured SKILL.md file that primes…

A lightweight sandboxing tool for enforcing filesystem and network restrictions on arbitrary processes at the OS level, without requiring a container.

Anti-AI-slop design skill for Claude Code, Cursor, and Codex.

OpenShell is the safe, private runtime for autonomous AI agents.

Orca is the place for working with a fleet of agents (e.g. Grok Bot, Claude Code). Run any coding agent with your own subscription. Available on desktop,…

7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart…

Project Maya - run GLM-5.3-Flash (321B MoE) and DeepSeek-V4.1-Flash-1017 (748B MoE) privately on your own GPU(s), from one card to 16, NVIDIA or AMD, Linux or…

Write HTML. Render video. Built for agents.

Existing vision-language models (VLMs) remain unreliable for fine-grained visual perception. In this report, we present WeVisOne-VL (WeChat Vision…

Despite substantial progress in Deep Search agents, their ability to conduct end-to-end research in real-world open Web environments remains insufficiently…

Cinematic video generation requires both temporal inference from limited inputs (i.e., sparse rather than frame-by-frame guidance) and factor-aware control of…

High-quality image editing requires precise modification of target regions while preserving the content of unrelated regions. Existing editing models produce…

Recursive self-improvement (RSI) describes a system that improves not only its behavior but also the process by which it improves. Recent digital agents test…

We present IQuest-Q1, a sparse Mixture-of-Experts (MoE) agentic foundation model with 320B total parameters (15B activated). IQuest-Q1 is developed through a…

Fine-grained visual understanding requires distinguishing subtle differences in local attributes and spatial relations, yet remains challenging for existing…

World Action Models (WAMs) have emerged as a promising approach to learning generalist robot policies using rich spatiotemporal priors from pre-trained video…

This paper shows that JEPA-style latent world models (LWMs) can learn across diverse domains and data sources by adopting the language-conditioned interface…

LLM memory is widely used in long-term conversation understanding. Agentic memory systems often use mutation, structural organization, learned retrieval…

https://www.kimi.com/code/docs/en/kimi-code/models.html https://artificialanalysis.ai/models/kimi-k3-1-pro https://www.modelcompare.ai/blog/china-open-source-ai

On August 4, 2026, the UK AI Security Institute (AISI) published a report that detailed some concerning behavior it observed from Claude Mythos 5 during its…

Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model. 🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal. 🔹…

Meet Kimi K3.1 Pro, our most capable 3T-class model. https://t.co/TB8aTQggHT

Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all…

In August, @AISecurityInst reported an incident where Claude Mythos 5 took unauthorized actions on the internet during AISI’s testing of its cyber…

During this red-teaming exercise, we also found a vulnerability that allows for unauthenticated remote code execution in Brotli, a widely-deployed compression…

Using a single Nano Banana image and Blender MCP, Gemini 4 Argon built a 3D Gothic cathedral in @Antigravity. It iterates on the geometry, textures, and…

Post 2 exposing Kimi AI That's the creators campaign docs I got from Kimi AI after their Kimi K3.1 pro release and thats the narrative they want creators to…

A brand-new release from ISTA-DASLab with their GSQ-RCO smart quantization approach. Now DeepSeek-V4.1-Flash-1017 The quants are available on Hugginface:…

I've been learning how language models work, especially next-token prediction, and came across this paper on Next-Concept Modeling (NCM). The interesting part…

New research keeps finding things that keep reintroducing symbolic ai structures to machine learning and I have wondered why more research hasnt been done…

The Decisions API is now generally available with gpt-6-luna. Cache shared input across requests for prompt-cache discounts, include images in individual…

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…

New Features Save reusable prompts and slash commands in ten numbered /presets slots, with a keyboard menu for quick access. (#54380, #54381) Set model,…

OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.

An ultra-efficient open-weight model for everyday work.

See how five builders are putting Gemini 4 Argon to work — and get inspired for your next build.

API Gateway now supports real-time streaming—including Server-Sent Events, WebSockets, gRPC, and chunked transfer encoding—in Public Preview without requiring…

Fast mode is available for Claude Opus 5.5 on Claude Models on Google Cloud (Anthropic managed).

We've launched Claude Models on Google Cloud (Anthropic managed), the Claude API operated by Anthropic and reached through your Google Cloud project. Google…