2026-10-11 AI 日报

Reverse engineer anything with agents, from app behavior down to native binaries.

AirLLM 70B inference with single 4GB GPU

Editorial diagram design for Claude Code, Codex, GitHub Copilot, Cursor, Factory Droid, and Pi. 44 diagram types. Self-contained HTML + SVG. No shadows. No…

Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the…

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…

YuE2: frontier music generation with symbolic planning, zero-shot covers, and agentic music editing.

Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

World's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent skill and production-knowledge files. Turn your AI…

High performance coding agent harness written in rust

The job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep…

Reinforcement learning (RL) has emerged as a key paradigm for improving large language model (LLM) agents, enabling stronger reasoning, planning, and tool-use…

Speech recognition on low-cost microcontrollers is, in practice, limited to keyword spotting or a fixed list of commands. We present Oído, an inference engine…

State-space models (SSMs) maintain a fixed-size recurrent state, but reading and writing it at every decoding step can bottleneck hybrid models. This eager…

While deep learning models for tabular data have recently received significant research attention, the choice of objective functions for regression tasks…

Regression-based RL for flow models mandates sampling \emph{without} classifier-free guidance (CFG) to align the rectification target with the reference…

Video reward optimization can increase proxy rewards while reducing video dynamics, a mismatch we identify as \emph{motion collapse}, including under a…

Tool-using agents increasingly require a common substrate for building, training, evaluating, distilling, and maintaining agentic systems, yet existing…

Gaussian process solvers for linear partial differential equations (PDEs) return posteriors rather than point estimates, but they condition only on equality…

Designing tool-integrated reasoning agent systems for complex chemical tasks remains a fundamental challenge. Although recent chemical agents have…

Time series causal discovery --- the process of detecting causation between time series --- has its roots in signal processing research from over a half…

Hi HN! I'm Saarth. I'm in 9th grade, and I built Atticus. Atticus is an incredibly fast and cheap Mac assistant that can do tasks for you. You hold down…

Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past…

Claude Next Model:) its over for image/video gen, if you are not impressed with this wait for two image gen output that will look that it was shot on camera…

I stumbled across this as I was currently having glm5.3 flash run only around 10tok/s basically unusable. With Maya i am running 30 tok/s now. GLM5.3 feels…

It's been 4 years since c.ai first hallucination model, and yet, we're nowhere good enough at LLMs in terms of spontaneity/interesting hallucination features.…

With the vLLM Semantic Router team we released Decision 3.0 (d3), our new SOTA decision model family ranging from 0.6B to 27B, Apache-2.0, all reading text,…

EDIT: since people keep calling it out, this is API key abuse but not exfiltration. Seems like both Harbor and Pier (sandboxes used in almost every Software…

Logs attached. We were running a DeepSWE variant on DeepSeek v4.1 Flash using the standard Pier sandbox. Out of all the 15 models we tested so far, ONLY…

I genuinely think this will be one of the biggest turning points in AI. Imagine having Claude Sonnet-level coding, reasoning, and agent capabilities running…

GPT‑6 is rolling out globally in ChatGPT with Intelligent UI, delivering faster responses with visuals and interactive experiences you can explore and use…

text-to-image

Using GPT-6 Astra in Codex, Asana made its browser agent 76x cheaper and 5x faster in tests to offer customers more capable models.

We've launched Claude Haiku 5.5 (claude-haiku-5-5), our most capable model tuned for high-volume and latency-sensitive work. It has a 1M token context window,…

automatic-speech-recognition

text-classification

OpenAI publishes new results on open problems in mathematics from an internal frontier model and shares Lean proof formalizations and research details on…

Mistral Large 4 (mistral-large-4) is now available in Public Preview. For details, see the Mistral Large 4 model card . Mistral Large 4 is an open-weight,…

Anthropic is committing $150 million over three years to the Genesis Mission, a federal initiative to accelerate scientific and technological discovery…

We’re launching the Anthropic Cyber Mission, a long-term commitment to securing the systems everyone depends on.