2026-09-20 AI 日报

A framework for building agentic apps

A straightforward method for training your LLM, from downloading data to generating text.

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode,…

Fault-tolerant, highly scalable GPU orchestration, and a machine learning framework designed for training models with billions to trillions of parameters

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining…

Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work.

Secure environments for developers and their agents

Let AI agents use your real, logged-in browser without interrupting your work. CLI + extension for browser automation across any shell-capable AI agent.

Omnidirectional video saliency prediction plays an important role in many immersive multimedia applications, including viewport-adaptive streaming and…

The gold standard datasets for mapping connectomes are electron microscopy volumes of densely labeled neural tissue at nanometer resolution. Yet…

Mixture-of-Experts (MoE) scales capacity, but existing designs cannot set three quantities independently. For a single token, participation is how many…

Comprehensive video understanding is crucial for advancing artificial intelligence toward the intricate dynamics of the physical world. While recent advances…

Robot localization is an ongoing challenge that demands mapping and positioning systems that are tolerant to viewpoint change. Event cameras are attracting…

Factual hallucination is commonly defined by incorrect factual outputs. We study a paraphrase-induced hallucination setting, where a model answers a factual…

Text-guided video editing modifies target content while preserving the appearance and temporal coherence of unedited regions. Training-based approaches…

How many human judgments does a panel of language models represent? The answer depends on what is matched. We audit categorical judge panels against empirical…

Autonomous software generation (ASG) aims to turn human requirements into executable applications, but delivering these applications does not necessarily…

Fairness research on language models involves measuring bias, applying mitigation methods, and examining the evidence on which an evaluation rests. Existing…

Sorry for the pretentious name, I know, I know.. It just contains all the pieces I would like to see a AGI model to have, and I can't stand the temptation.…

Meet Qwen-Image-2.1, the most balanced and cost-effective image generation model in the Qwen-Image series! Now open weights! 🎨 A unified model for both…

Qwen-Image-2.1 is now supported in ComfyUI! Try it now and share your creations! 🖼️

We’ve received so much love for Qwen-Image-2.1 over the past 24 hours, thank you!! Also gotten a lot of questions about the license, especially around model…

ZCode is now open source, and the reported security issues have been addressed. Source code: https://github.com/zai-org/ZCode The repo includes its desktop…

Meet Qwen-Image-2.1, the most balanced and cost-effective image generation model in the Qwen-Image series! Now open weights! 🎨 A unified model for both…

But then how should we evaluate its “intelligence”? Does Jev have benchmarks comparable to the usual LLM benchmarks, or is the comparison fundamentally…

Instead of generating text, it takes agent state plus available actions and directly returns action probabilities. Architecture: state: frozen Qwen3 0.6B…

We are excited to open-source Qwen-Image-2.1, an image model in the Qwen family that balances generation quality, inference efficiency, and cost.…

text-to-image

Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking are our most advanced live dialogue models yet, built for natural conversation.

Simultaneous interpretation is not only about translating fast — it must also hear clearly and translate accurately. Qwen3.8-LiveTranslate rebuilds real-time…

Z.ai's coding agent harness. Powerful, intelligent, extensible.

The Compliance API local session endpoints now also return transcripts of Claude in Chrome sessions (product_surface value claude_in_chrome), in beta for…

For cache diagnostics, a response to a request that sends the cache-diagnosis-2026-04-07 beta header now always includes the diagnostics field. The field is…

Google has partnered with Speakeasy to open-source their OpenAPI code generation suite under the AGPLv3 license, a strategic move prompted by the sudden…