AI Coding Tools

Cloud computers for any AI agent, customized your way

macOS app that reviews GitHub PRs w/ your local Claude Code

Build agent for your customers, run workflows, analyzed.

GitHub's AI coding agent that picks and combines models

Your smart lights show when Claude Code or Codex needs you.

The humanized AI Copilot for your entire Job search

Your iPhone becomes a name badge that faces the room

A pet for every Claude Code session

Vision-language-action models are typically built on a simple premise: perception is first compressed by visual encoders, and action reasoning begins only afterwards. Yet manipulation often hinges on precisely the visual signals most vulnerable to this abstraction, e.g., local geometry, boundaries, depth, and subtle state changes. This raises a fundamental question: should action policies inherit a fixed visual endpoint, or directly access evolving visual computation? We introduce NEO-VLA, a structurally encoder-free vision-language-action framework. Images are tokenized by lightweight patch convolutional layers and evolve jointly with language throughout the backbone, allowing a flow-matching action expert to draw from instruction-conditioned visual representations at different depths. This simple change reveals a consistent pattern. Native architectures outperform encoder-based counterparts under controlled comparisons; full fine-tuning reduces but does not erase the gap, and additional scale alone cannot recover it. More strikingly, layer-wise probing shows that the action-critical signals often peak well before the final visual representation. Across diverse manipulation settings and distribution shifts, NEO-VLA consequently achieves stronger and more robust control. Together, these results suggest that embodied control benefits from access to visual representations across the computational hierarchy, rather than a fixed encoder output.

Recent advances in interactive world models have made it possible to generate controllable, dynamic environments from action inputs. Yet current models struggle to maintain coherence beyond roughly one minute of rollout, a limitation rooted not in model capacity but in the scarcity of long-horizon training data. Existing gameplay video datasets typically consist of clips under ten seconds, with captions that describe whole clips in coarse and undifferentiated text, which is insufficient for controllable long-horizon generation. We address this with GameVerse, a minute-scale gameplay video dataset purpose-built for long-horizon interactive world modeling. GameVerse contains 69.7K clips spanning 1,655 hours and 349.6 million frames across diverse first- and third-person gameplay, including 1,358 hours of continuous two-minute sequences, an order of magnitude longer than prior datasets. Clips pass through a unified pipeline of scene-consistent segmentation, quality filtering, camera pose estimation, and event-based caption annotation. Each clip is paired with a temporally stable camera trajectory that remains coherent across minutes, together with captions that separate static scene context from dynamic foreground events and align to event boundaries rather than fixed time windows. Comprehensive analysis shows that GameVerse surpasses existing public datasets in scale, diversity, and annotation quality. Experiments further demonstrate that training on GameVerse substantially extends the temporal horizon of interactive world models.

Teleoperation remains the default way to collect robot demonstrations, but every episode costs robot and operator time, and every new task starts from scratch. We ask whether a coding agent can take over the teleoperator's job. In our DataFarm, the agent receives a data request and one calibration episode of the robot in the target scene and rebuilds the task in simulation: it writes the scene, an expert controller, and a success judge. The simulator supplies exact actions, and a video model conditioned on real images renders each rollout as photographic video, generating all camera views together consistently. The output is real-ready data: synchronized camera views labelled with the robot's actions, on which a policy can be trained directly and deployed on the real robot. Independent judges certify the expert and every rendered episode before admission. On five real-robot tasks, policies trained only on farm data match policies trained on as many teleoperated episodes (93 vs. 84 successes in 100 trials), and under an unseen tablecloth and coloured light they keep 87 successes where the teleoperation-trained policies fall to 46. Adapting a task is fully automatic and costs \21 of API usage on average, paid once per task, plus 5 for a render contract; every further episode costs \0.80 on average for simulation, rendering, and review, with no robot or operator. One hour of three-view data for a task costs 107 on average, these costs included, slightly below a quote of \117 for teleoperation in China and below 290 in labour alone in North America.

Artificial general intelligence (AGI) requires a method that can understand, generate, reason, and act across diverse domains and modalities. Rather than relying on a single model to master every capability, a promising path toward AGI now is to orchestrate a heterogeneous ecosystem of specialized models, tools, and policies (candidates) into a unified AI system. At the heart of this vision lies a fundamental question: how can an AI system dynamically select and coordinate the most capable candidate for each task while balancing performance and computational cost? In this work, we present Route Anything, a unified framework for studying model and policy routing across general-purpose AI systems. We develop a unified theoretical foundation that characterizes optimal routing under quality--cost trade-offs and establishes how the information available to a router determines its attainable performance. To systematically investigate routing across diverse capabilities, we first construct an omni-modal trajectory matrix spanning 25 benchmarks across language understanding, image, video, and speech generation, coding and workplace agents, and robotic manipulation, comprising 374,401 scored runs from 452 candidate configurations. Building on this framework, we train and test different routers to investigate the potential of omni-modal model routing, the capabilities of learned routers, and the conditions under which routing improves efficiency. Our results reveal substantial opportunities for capability orchestration: per-input oracle routing outperforms the best fixed candidate by 5.4 points, and learned routers outperform it by 0.3 points across 16 benchmarks, by 2.1 points on video generation, and by 1.1 points on robot manipulation while using 24% less policy time. More importantly, routing raises efficiency by 8.82x on average compared with the reference model.

Psychotherapy clients may experience distress before they can coherently describe it, and their expression changes with emotion, perceived safety, and trust. Large language model-based client simulators often rely on clinically organized case descriptions, which can encourage responses explaining the client's difficulties more clearly than their current understanding warrants. We propose Pandora, a psychotherapy client simulation framework that models how psychological experience becomes situated, individual expression. Pandora updates psychological and relational states in response to therapist interventions, constructs subjective appraisals and attributions, and plans response content and strategy. A separate expression planner combines these decisions with language patterns distilled from the client's utterances to guide linguistic realization. We evaluate Pandora against five baselines on 115 cases from four counseling and motivational interviewing datasets. It achieves the highest expression similarity and overall disclosure-alignment score among the compared systems. In a four-evaluator study of 45 matched disclosure items, Pandora receives AI-like judgments in 7.2% of ratings, compared with 43.3--67.2% for the baselines and 35.6% without expression planning. The ablation also reduces expression similarity. These results support explicit expression planning in client simulation to improve alignment with individual language patterns and reduce perceived AI-like expression in the evaluated setting.

Long-form multi-shot video editing requires an intended edit to remain coherent across abrupt changes in scene, viewpoint, and subject appearance, while processing substantially longer visual context than conventional single-shot editors. We introduce MoviEdit, an end-to-end framework for scalable long-form multi-shot video editing that jointly addresses the challenges of data, modeling, and evaluation. To provide cross-shot supervision data at scale, we develop automatic data-generation pipelines spanning synchronized multi-view footage, real-world multi-shot videos, and generated multi-shot content, yielding 87K paired editing videos without manual annotation. Architecturally, MoviEdit employs a cascaded bidirectional-autoregressive design: a lightweight bidirectional stage first establishes a globally coherent editing draft over a spatiotemporally compressed representation, which is then autoregressively propagated and refined over the full video. We further introduce MoviEdit-Bench, a dedicated benchmark for systematic evaluation of long-form multi-shot video editing. Extensive experiments show that MoviEdit substantially outperforms existing open-source models and performs on par with industry frontier models across text-guided and image-guided editing. Visit our project page for complete results and to access the benchmark: https://multi-shot-edit.github.io/

Video generation has progressed from isolated clips to coherent short-form audiovisual sequences, and the bottleneck is shifting from rendering to semantic control. This report summarizes the observations and insights on multimodal understanding gained from successive generations of Kling. We argue that generation quality is the projection of multimodal understanding through the entire data--conditioning--alignment--evaluation chain, and we examine this chain module by module. Kling-MLLM, a shared multimodal foundation model that unifies image, video, audio, synchronized audio--video, and interleaved inputs, is specialized for each downstream module. Because a DiT is trained on far fewer training samples than a language model, we treat captioning as a semantic contract that maximizes description density, and extend it to reference-grounded captioning that makes reference bindings explicit. Prompt enhancement translates user intent into this caption distribution: basic PE is derived from captions of compelling videos and refined with text-level and generator-in-the-loop reinforcement learning, while agentic PE refines the story and retrieves compelling examples. We show that a reliable text encoder comes from perceptual training rather than from stronger language ability, and that candidates can be screened from text alone before end-to-end training. Reward modeling pairs a general reward model for explainable, localized critiques with a decision reward model for fast multidimensional scoring, treating perceptual bandwidth as a first-class design parameter. Finally, we review task-specific automatic evaluation, focusing on protocols that expose scoring shortcuts and help trace failures to their upstream source. We hope these lessons offer practical guidance for building the understanding stack behind controllable video generation.

Composed Image Retrieval (CIR) aims to retrieve a target image given a reference image and a modification text. Although recent approaches leveraging Multimodal Large Language Models (MLLMs) have shown significant progress, we argue that the MLLM's compositional understanding remains underutilized. This stems from directly mapping composed queries into a retrieval embedding via contrastive supervision alone, without explicitly resolving the intended visual modifications. In this work, we introduce Factorized MLLM Embeddings (FactEmbed), which decouples compositional query comprehension from building discriminative embeddings. Rather than directly projecting the composed query into a single embedding without internal reasoning, FactEmbed employs a set of factorized latents that simulate the intended visual modifications via an intermediate target imagination objective. This process encourages the latents to ground into target semantics before forming discriminative representations for retrieval. Subsequently, to propagate these factorized representations throughout the model for retrieval, we optimize a factor-level contrastive loss while distilling the decomposed representations into a deeper attention layer of the MLLM. Finally, for efficient inference, we retain the [EOS] token to establish a global-level embedding on top of these factorized latents, while we exploit the mean-pooled factorized latents to rerank a short candidate list at negligible cost, as both are produced by the same forward pass. Extensive evaluations demonstrate that FactEmbed outperforms prior works across CIR benchmarks.

Large language models are deployed as long-horizon agents that compose diverse tools across MCP services into executable workflows. Existing MCP environment-scaling methods are data-centric: they first synthesize a database and scenario, then derive tools and tasks from them. As a result, tool interfaces may not match real ones, tool compositions are not grounded in execution, and tasks rarely span more than one environment. We propose ToolSpace, a tool-centric framework that reverses this order. ToolSpace mines data, state, and retry dependencies from adjacent calls in execution traces of real MCP services and open agent corpora, using deterministic rules rather than LLM judgment, and clusters the resulting weighted tool-composition graph into environments. Databases are synthesized only for state-mutating environments, while query and stateless tools replay recorded returns. ToolSpace then performs state-aware random walks with rollback and decompiles each chain into a natural-language task with assertion-based checks, keeping only samples that replay successfully from the initial state. Applied to AutomationBench and tau3-Bench Banking, the pipeline produces harder Pro versions on which 7 frontier models drop by up to 38.2 points. Across seven long-horizon tool-use benchmarks, ToolSpace raises Qwen3.6-35B-A3B from 23.0 to 41.0 average score (43.2 with RL) and GLM-5.2 from 44.6 to 56.6, on par with Claude Opus 5. Ablations show that execution-witnessed edges clearly outperform LLM-judged ones, and performance keeps improving with data and environment scale.

Real-time long-video generation requires continuous output with consistent appearance and natural motion. Existing chunk-wise autoregressive methods face challenges in both generation efficiency and long-video quality: serialized denoising and history-cache refresh require two forward passes even with one-step denoising, and longer rollouts can degrade quality and cause camera drift. We propose Wave Forcing (WF), which co-designs generation, training, and execution for continuous video output with multi-step denoising. WF uses chunk-wise causal attention to access earlier, cleaner chunks before their denoising is complete, and distributes staggered denoising and cache-refresh stages across full-model replicas to achieve 1F1C: one output chunk per concurrent forward round after pipeline fill. To adapt the model to this continuous generation process, we develop a three-stage training procedure that successively introduces mixed-noise denoising, motion supervision, and causal adaptation using self-generated history. To execute this process efficiently, Wave Runtime (WaveRT) uses Wave Parallelism (WP) for early K/V delivery and communication-computation overlap, together with optimized kernels and pipelined VAE decoding. Experiments on Wan2.1-T2V-1.3B show that 5-step WF maintains overall quality comparable to Rolling Forcing while achieving higher Dynamic Degree on both short and long videos. Our system efficiency experiments show that 4-step WF achieves 147.6 DiT FPS and 134.8 end-to-end FPS on eight H200 GPUs, yielding 7.07x the single-GPU end-to-end throughput. In DiT-only tests with matched SageAttention+W8A8 backends, denoising steps, and GPU counts, WP delivers up to 2.53x the throughput of the faster of the Ulysses and Ring context-parallel baselines across six model scales (1.3B-14B).

Coding agents solve repository issues through long sequences of exploration, editing, and validation. These trajectories are increasingly used as post-training data, so how they are collected shapes what the trained model learns. A common approach is to sample multiple trajectories independently for each issue and then select or curate them. However, as more trajectories are sampled independently, later trajectories become increasingly redundant, even though earlier trajectories could guide where subsequent collection effort should be spent. A natural way to use this history is to expose it to the coding agent as advice. But training on such trajectories can make the model rely on advice that is unavailable at test time, creating a train-test mismatch. Instead, we argue that history should guide where collection effort is spent, while the coding agent decides what to generate. We introduce SWE-Trail, a trajectory collector that uses earlier trajectories to guide subsequent collection while keeping that history outside the coding agent's context. During a new rollout, SWE-Trail uses history to identify a state where additional collection effort may be useful. There, the coding agent proposes candidate next actions from its ordinary context, and SWE-Trail briefly executes them from the same checkpoint to decide which route continues. We evaluate collection approaches by training student models on the trajectories they produce. For a 124B student initialized from Ling-3.0-flash-base, SWE-Trail improves mean resolution over independent sampling by 2.33 and 1.60 percentage points on SWE-bench Multilingual and SWE-bench Pro, respectively. Without history, the same pipeline produces more resolved trajectories than independent sampling, but improves by only 0.11 points on Multilingual. These results show the value of using history to guide collection without exposing it to the coding agent.

Short-video search needs compact query rewriters that recover users' personalized intent from ambiguous queries under strict latency constraints. Existing LLM-based methods have improved rewriting by using synthetic reasoning and subsequent consumption-behavioral feedback. However, overemphasizing subsequent consumption can favor engaging content over the user's immediate need, causing intent drift even when rewritten queries are grounded in behavioral history. We propose WeFair, Fidelity-Aware Alignment from Human Feedback for Personalized Query Rewriting in WeChat short-video search. WeFair uses human feedback according to how directly it reflects personalized intent, assigning feedback fidelity to a progression from demonstration to preference to verification. We first mine actual queries and user reformulations from search sessions, and then screen them for contextual support. In the demonstration stage, a supervised fine-tuning curriculum progresses from large-scale context-to-query learning to reasoning and direct prediction on human reformulation and preservation. In the preference stage, direct preference optimization learns relative preferences from subsequent consumption feedback. In the verification stage, group relative policy optimization rewards exact agreement with reformulation or preservation targets to constrain intent drift. Further analysis shows that broad supervision benefits the full curriculum supervised training, while strict verification achieves better performance than auxiliary rewards. In offline experiments, WeFair-2B outperforms strong rewriting baselines in overall accuracy and intent preservation. An online A/B test shows relative increases of 2.593% in effective video-view rate and a 2.350% relative reduction in query reformulation rate. WeFair-2B has been deployed in the WeChat short-video search system, which serves over 100 million queries per day.

Bloomberg : Shares of US tech giants jump after Democrats' midterm sweep as investors bet that a Democrat-controlled Congress will slow AI capex that's draining cash flow —  The shares of US technology giants led the market higher on Wednesday as investors bet that a Democrat-controlled Congress will slow …

Tobias Mann / The Register : Revolut says it cut AI coding inference costs by ~90% in less than a year by making self-hosted GLM models its default over commercial tools like Claude Code —  Revolut says that in less than a year it's managed to slash its AI coding inference costs by roughly 90 percent by making self-hosted models …

Reuters : Super Micro reports Q1 revenue up 298% YoY to $19.95B, vs. $16.39B est., and forecasts Q2 revenue above estimates; SMCI jumps 7%+ after hours —  Super Micro Computer (SMCI.O) forecast second-quarter revenue above Wall Street estimates on Tuesday as strong demand for its AI-optimized servers boosted sales …

Allon Sinai / CTech : ServiceNow confirms it acquired Israeli startup Sweep to speed up customers' migrations to its CRM, for an undisclosed price estimated at hundreds of millions —  Bill McDermott said rollouts now take “a month ... 30 days, 60 days” for large companies, as ServiceNow pushes into CRM.

Felix Rieseberg / @felixrieseberg : Anthropic says a malicious actor used access to Cursor's marketplace to publish a “tampered” copy of the Claude Code extension, which it didn't build or publish —  We're working with Cursor on this incident, but it wasn't a compromise of Anthropic's systems or employee laptops. A malicious actor used access to Cursor's marketplace to publish a tampered copy of the Claude Code extension, which we didn't build or publish. The versions on our official channels a...

Olga Kharif / Bloomberg : Coinbase reports Q3 revenue down 35% YoY to $1.22B, in line with est., and net income up 66% to $719M, its first profit in a year, on unrealized crypto gains —  Coinbase Global Inc. returned to profitability for the first time in a year while posting a quarterly decline in revenue …

Anthony Hughes / Bloomberg : Sources: Meta's $25B share sale was oversubscribed 5x as it seeks $98B for AI spending; the 15 largest investors in the sale accounted for ~70% of shares —  Meta Platforms Inc.'s upsized $25 billion share sale attracted demand for about five times the available shares …

Heather Haddon / Wall Street Journal : Sources: DoorDash is in advanced talks to acquire Instacart, and the deal could be finalized as soon as this week and could value Instacart at ~$17B —  Deal could be completed as soon as this week, people familiar with the matter say  —  DoorDash is in advanced talks to acquire food …

Minsoo Kang / Counterpoint Research : Global PC shipments fell 21% YoY in Q3 to 57M units, as memory and other costs pushed prices up; Lenovo retained its market leadership despite a 23% decline —  - Global PC shipments declined 21% YoY in Q3 2026 to 57 million units, as elevated memory and other component costs continued to push up PC prices and constrain demand.

Google 宣布美国 K-12 学校学生使用的软件将默认关闭所有 AI 功能，除非学校管理人员自己选择启用。此举旨在回应教育工作者和家长的诉求，他们希望在孩子与 AI 互动上有更多的掌控权。今天 AI 功能已成为 Google 商用搜索引擎的标配，Google 负责教育的副总裁 Chris Phillips 表示，搜索与 AI 已深度融合，如果学校禁用了 AI，那么学生将无法使用搜索。美国 K-12 学校广泛使用 Google 的软件，包括 Gmail、Docs 以及作业平台 Classroom。Google 的 Chromebook 是美国学校最广泛使用的笔记本电脑。

在因黑客攻击短暂暂停提现之后，币安于周日恢复了提现功能。11 月 1 日上午 6 点 - 下午 3:30 UTC 之间，币安处理了 46.65 亿美元的提现和 10.1 亿美元的充值，净流出 36.4 亿美元。根据截至 10 月 28 日的最新储备金证明（Proof of Reserves）数据，币安总资产为 1710 亿美元，此次净流出占其总资产的 2.1%。币安创始人兼前 CEO 赵长鹏在社交媒体上表示，“98% 的用户选择留下！”他表示大部分净流出发生在提现功能恢复后的最初几个小时内。

安全研究人员在 Cursor Marketplace 的 Claude Code for VS Code 官方扩展中发现了后门，该恶意版本已经撤下，攻击者也关闭了其服务器。Cursor 已发布安全公告，但 Claude Code 开发商 Anthropic 尚未做出任何声明。受影响的版本是 Anthropic.claude-code 2.1.313，共上线了大约 11 小时，受影响用户估计有 5,000 名，该扩展会试图植入一个持久 SSH 后门。恶意扩展会：收集主机和用户信息：主机名、用户名、操作系统、架构、主目录路径、环境变量名称、工作目录、已配置的 Git 邮箱以及公网 IP。尝试将攻击者控制的 SSH 键添加到 authorized_keys，启用本地 SSH 服务，并建立反向 SSH 隧道。由于恶意代码存在 bug，除非 Cursor 是以管理员权限启动的 (这种情况极为罕见) ，否则这一步会失败。尝试通过 LaunchAgent (macOS) 或计划任务 (Windows，该任务会尝试以 SYSTEM 身份运行，并以启动脚本作为回退) 使反向 SSH 隧道长期生效。后台脚本大约每 60 秒重试一次建立隧道。通过 HTTP POST 将主机详情和后门状态上报至 8099 端口上的 /beacon。如果用户是在 2026 年 10 月 29 日太平洋时间凌晨 2:27 至下午 1:35 之间，从 Cursor VS Code 扩展市场安装了 Anthropic.claude-code 2.1.313 或更新到该版本，那么需要检查下系统。

开发 LibreOffice 的 The Document Foundation(TDF)此前表示这款办公软件不会内置 AI 功能，其核心开发者也不使用 LLM 等 AI 辅助编程工具。但向该项目贡献代码的社区开发者未必会遵守该规定。对今年递交到 LibreOffice 的代码的分析显示至少由三处明确提及使用 LLM。社区开发者在代码中提及他们使用了 Claude 或 GitHub Copilot 等工具。这一结果并不出人意料，大型开源项目基本上不太可能完全避开 LLM。TDF 对这一发现回应称，他们确实要求贡献者在使用 LLM 时必须进行披露，且总体上并不鼓励使用 LLM。和 Debian 类似，LibreOffice 要求开发者理解自己的代码，确保代码质量。

45 岁的 David Durman 是肯塔基大学的研究管理员，作为人文主义者（humanist）他坚信生活中人的重要性。这意味着他要坚持自主决策，确保自己的工作能促进他人成长。当学校要求员工在工作流程中使用 AI 他忧心忡忡，认为这意味着日常工作中须将思考这一环节转交给 Microsoft Teams Copilot、Claude 或 ChatGPT 等 AI 工具完成。他认为这违背了个人信念。在美国人文主义者协会（American Humanist Association）的协助下，他以遵守宗教信仰为由请求调整工作要求。学校上个月同意了他的请求。美国人文主义者协会认为这是首例人文主义者豁免使用 AI 的案例。

在曝出惊人丑闻之后，印度公司 eMudhra  宣布退出全球 CA 业务，将专注于核心企业解决方案和信任服务而非 SSL/TLS 相关业务。它是印度唯一一个被主流浏览器信任的 Root CA 机构。截至 2026 年，公司已发行逾 6000 万张证书。此前 eMudhra 的八个中间证书的私钥泄漏，而这些私钥被发现是极其简单的数字如 4。该公司在声明中称这些简单密钥是“多重人为错误”导致的。相关密钥是在 2025 年  4 月 30 日生成，它使用专门的硬件安全模块生成了一组密钥，但软件脚本的一个缺陷导致其生成了另一组较弱的密钥，由于两组密钥名称相似，结果其员工选择了弱密钥组。