2026-09-29 AI 日报

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

📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG

Multi-agent harness that runs Claude Code and Codex together as one system

25 MB lightweight cross-platform database client for 100+ databases, including MySQL, PostgreSQL, SQLite, Redis, MongoDB, DuckDB, SQL Server, and Dameng.…

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

🔥 Supercharge your AI agents with data from the web and beyond. A web data API to search, scrape, and access more sources.

Learn it. Build it. Ship it for others.

Independent Auditing of AI Agents. Run by human or the agent itself, to answer the most crucial question in the AI Agent Economy. Is the agent doing what is…

A smarter, self-hosted AI assistant — multi-user, multi-agent.

Skills for Real Engineers. Straight from my .agents directory.

Symbolic models make melody, harmony, rhythm, and form explicit but typically stop before a finished recording; audio models produce complete songs while…

FullAttn often assigns negligible normalized mass to much of the causal score space, yet dense kernels execute the complete post-score path after forming each…

We find that language models can transfer capabilities through task-unrelated text. Post-training typically improves language models using task-specific data.…

Vision-language-action (VLA) and world-action (WAM) models map observations and instructions directly to robot actions. This directness ties a policy to…

Reinforcement learning can turn one language model into several specialists, each excellent at a single skill such as mathematics, coding or following…

Looped transformers have demonstrated promising parameter efficiency by reusing layers for latent computation. Prior studies compare looped and non-looped…

An agent can complete a task while exhibiting undesirable behavior during execution. Developers need tests for the specific behaviors encountered in…

Reinforcement learning (RL) for code agents often uses executable tests to provide binary rewards. With these rewards, Group Relative Policy Optimization…

Unified multimodal models can both look at and render images, so in principle they can repair their own generations: diagnose what an image gets wrong, revise…

Real-time full-duplex speech models can listen while speaking, enabling natural interaction without rigid turn boundaries. Existing benchmarks evaluate…

Andrew Fasano, Marius Fleischer Cole McFaul, Robert Xiao, Tripp Gallagher Five months ago, we announced Claude Mythos Preview, the first AI model that could…

Ultrafast is our fastest way to build with Astra yet: in Codex, it runs up to 8x faster than Astra Standard and 4x faster than Astra Fast. Bring your ideas to…

Like.. yea bro, I knew GLM was cool. Now everyone does.

"Near-Astra intelligence for a fifth of the price"

What do you want from AI? We’re launching a new study with Anthropic Interviewer to learn more about your experiences using AI, what role you want it to play…

I think most people are sleeping on this inference engine. I tried multiple llama.cpp forks and none of them comes close to the inference speed of Strata.…

I’ve spent the last few months writing something I wish I had when I started working on ML performance engineering. It’s called How to Make Your Model Fast: A…

We have a small feature that tags incoming support tickets before a person sees them, mostly deciding whether something is a billing problem or a bug report.…

I've been running Claude Code, Codex, ohmypi, and a couple others locally and it kind of bugged me that I had no real idea what they were doing outside the…

No cameras were harmed in the making of this Odyssey pilot.

Meet GPT-6.1 Sol: near-Astra intelligence for coding, computer use, and professional work at one-fifth of Astra’s standard API input and output token prices.

Dots by OpenAI are proactive assistants that can keep working across complex projects and everyday tasks. Learn how dots help you stay in control while work…

Like Claude Mythos Preview, GLM-5.3 has strong capabilities for autonomously building end-to-end cyber exploits. But GLM-5.3 is unlike other frontier models…

Z.ai GLM 5.3 (zai-glm-5-3) is now Generally Available.

An internal component used by DeepSeek Harness

An internal component used by DeepSeek Harness

We’re launching a new study using Anthropic Interviewer to learn from your experiences with AI, and we invite you to participate.

Explore more than 20 announcements from OpenAI DevDay 2026, including GPT-6 Astra, ChatGPT, Codex, APIs, security, and new tools for builders.

A high-performance communication library for machine learning training and inference on Huawei Ascend NPUs.