Baidu introduces a promotional price of 4.9 yuan / 1000 searches for the Search AI API, lower than DeepSeek's 5 yuan / 1000 searches. Wenxin Assistant shifts to 'one-time delivery', AI lens upgraded to 'flowing' tracking, and adds 9 new dialect conversation modes. The number of GUI interactive scenarios is increased 5 times.
Baidu announced a full upgrade to its search AI API, covering over 170 scenarios; the GUI interactive scenario on the search results page has increased by 5 times, and it has integrated more than 50 billion data resources from government, livelihood, and the People's Bank of China.
Vioooza and overclocker Clemens "CENS" Nüssler jointly launch TG x CENS LN2 Container Liquid Nitrogen Cannon, designed for GPU extreme overclocking, compatible with AMD and NVIDIA graphics cards, using ultra-thin nickel-plated copper material and spring-loaded system to achieve precise temperature control at -196°C, priced at 599.90 euros including tax
DeepSeek official website code exposed, Cyber security model application entry, including cyber-personal-v3 (personal version) and cyber-enterprise-v4 (enterprise version) two sets of application templates, personal version requires ID card face verification, enterprise version requires business license authorization letter, currently only small-scale gray testing.
Transformers releases66GitHub Copilot released v5.22.1 patch version, including the latest feature improvements and fixes.
Douyin Work releases the configuration of third-party large model services for enterprise subscribers on the same day, the first batch includes Kimi 3.1 Pro, GLM 5.3, DeepSeek V4.1 Flash and other series models, the function will be enabled by default after launch, enterprises do not need to purchase or apply additionally, subsequent usage will deduct AI points accordingly.
Xiaomi REDMI's official response: The K100 uses the Snapdragon 8 Ultra edition, which is not the latest platform, but some game tests show it does not lag behind the Snapdragon 8 Gen 6 Pro. It is equipped with the AI dedicated graphics card D2, supporting 100+ games with 165Hz frame rate. The REDMI Clip Neo adopts a four-microphone design, paired with AI call noise reduction algorithms to optimize wind noise and other background noises.
Hisense DaSheng G5 PRO AI Monitor Listed on JD.com, Equipped with 2K 200Hz Fast IPS Panel, GtG Response 0.4ms, Price 1299 Yuan, 999 Yuan After National Subsidy in Some Regions. Supports 10-Bit Color, Provides 2 HDMI 2.1, 1 DP 1.4 Port.
Baidu has open-sourced the Wenxin Unlimited OCR 2 model, which supports parsing documents totaling over 3,600 pages at once, natively supporting layout element segmentation recognition and cross-page text merging, table recognition and merging. It scored 84.6 on the MPDocBench benchmark test and 77.1 on the MarathonDocBench benchmark test.
Spotify plans to launch its AI music mixing service in the first quarter of 2027, allowing paid subscribers to use artificial intelligence to reconfigure or cover existing songs, and considering a monthly additional fee of $4 to $5. The company is aiming to release the service around the Grammy Awards ceremony next February, but the specific launch time and pricing have not been finalized.
STEPX Neo, the native intelligent agent smartphone under the leap terminal, welcomes its first upgrade two weeks after launch, with updates covering video creation, life services, work tasks, and optimization of the underlying experience. After the upgrade, users can call Jianying to generate Vlogs, create picture and video remixes, and receive proactive creative suggestions; they can also actively provide Alipay coupons, place orders through Alipay AI Pay, or assign complex tasks to WorkBuddy to complete. The continuous support of ecological partners reflects the industry's recognition of the leap large model native route.
Zhimi LS8 NEO pure electric version pre-sales, length, width and height 5085/2000/1807mm, wheelbase 3060mm, turning radius 4.85 meters, the entire system is equipped with bidirectional 24° intelligent rear wheel steering and a three-fold safety redundancy line control chassis, equipped with CATL 93 degree/103 degree batteries (CLTC range 684/752 kilometers), 15-minute fast charging 529 kilometers, IM Claw Lobster intelligent agent first non-commercial owner Token lifetime free.
Meta has launched the AI fashion app Style Sketch, where users can link their Instagram accounts to generate a virtual wardrobe and receive outfit recommendations, and can also try on clothes on a virtual model. The app supports iOS and Android systems and has already launched in Indonesia, Singapore, and Thailand.
vivo Y600 phone launches today, featuring a 10000mAh large capacity battery, supporting working temperatures from about -35℃ to 50℃, priced from 1699 yuan.
iPhone version Claude voice mode now supports Apple CarPlay, allowing users to ask Claude directly while driving; Anthropic also announced that iPhone and iPad versions will support pull-down search function and allow multiple accounts to be logged in and switched simultaneously.
The QQ HarmonyOS version App 9.2.80 (3570) beta upgrade has been launched, adding support for a simple mode, viewing space post views, enhanced group notification experience, selecting parts of plain text messages, AI repair for space albums, timed group mute and individual member mute, and chat number recognition features.
Google's AI programming assistant Jules, launched less than two years ago, will gradually stop services, its experience will be applied to the Antigravity platform. Jules is positioned to complete programming tasks that developers are unwilling to do, while Antigravity is a 'agent-first' development platform launched by DeepMind in November 2025.
Xinyan Robotics releases Embodied Family Intelligent Agent 'Bubbo 1', equipped with 13 types of 26 sensors, 6 degrees of freedom robotic arm, and dual expressive screens, capable of recognizing family members and analyzing heart rate changes. This intelligent agent is based on Bubbo OS to build Family Agent and has collaborated with MeeSoo Smart Ring to achieve data sharing.
Claude Code releases1515Claude Code v2.1.314 released, added write permission interception outside the working directory, optional identity group upstream routing, Bedrock upstream guard logs, idle conversation summary feature, fixed multiple session interruption, model switching, TLS connection, and tool call issues.
Google Gemini now supports directly generating, previewing, and editing PDFs, Google Docs, tables, presentations, Markdown, TXT, and code files in the chat. PDF editing functionality is supported on both web and mobile, while other formats are limited to the web. Previously, it already had file generation and download capabilities. This upgrade allows users to modify files without exiting the app and supports full-screen editing, text questioning, and content modification.
Gemini CLI releases1010Gemini CLI v0.67.0-nightly.20261030.gb633c0703 released, fixes case-insensitive isWithinRoot issue on Windows, identifies Gemini 3 models with dots, adds extension validation command and evaluation tests, handles leading slashes in file search queries, surfaces 429 rate limit cooling and countdown
OpenAI increases ChatGPT Pro and Business Premium subscription limits for Work and Codex usage, doubling Pro 100, Pro 200, and Pro 500, and cancels the original plan to reduce Pro 200 limits.
SGLang releases00GitHub Copilot v0.5.23 released, providing code generation, review, security, and collaboration features, supporting custom integrations and advanced security features.
OpenRouter Blog5858Claude Code installation is free, the cost depends on the model called. On OpenRouter, using an API key, the median cost for 2-9 rounds of Claude Opus 5.5 is 0.31 USD, and 10-49 rounds is 1.39 USD; the corresponding cost for Claude Sonnet 5.5 is 0.16 USD and 0.60 USD.
iOS 27.2 introduces 'When my iPhone opens or closes' automation trigger conditions, supporting hinge actions on foldable devices to trigger operations, such as automatically opening an app or starting Focus Mode when the phone is unfolded, or turning off the screen or pausing tasks when it is closed. Shortcuts extension 'Workout' automation is supported, allowing automatic recording or playing of playlists when exercise begins.
OpenRouter BlogOpenRouter Guardrails reconfiguration allows administrators to set organization-wide limits, providing different budgets and model access permissions for different people, and divides budget, model access, and content moderation into three independent policies. Organization administrators can set global budgets, zero data retention requirements, or available models and providers, and workspace administrators set defaults within this range.
Databricks Blog4747Databricks announces the general availability of Spark Declarative Pipelines Real-Time Mode, which achieves millisecond-level end-to-end latency through continuous streaming, parallel query stages, and instant data delivery.
Simon Willison6565OpenAI has launched a new offer for ChatGPT subscribers who meet the criteria, with $100, $200, and $500 monthly subscription plans eligible for an equal number of API credit limits. Previously, Anthropic has provided similar benefits for Max and Team subscriptions.
Factory NewsFeaturedSame story · 2 outlets7878Factory releases Droid 1.0, a major update to its intelligent agent harness, with programmatic tool calling enabled by default (Script mode). The model can execute hundreds of tool calls within a single request in a sandbox, executing JavaScript.
How the accounts differ:Factory News states the model can execute hundreds of tool calls in one request; MarkTechPost mentions the model request drives hundreds of tool calls.
Why it matters: The original text provides cost and time data for Droid 1.0 compared to Codex, Claude Code, and other harnesses on Terminal-Bench 4, which can be used to determine the selection of agent harnesses.
LangChain BlogFeatured6767LangChain adds browser capabilities to Managed Deep Agents, allowing developers to export browser configurations through define_browser, enabling managed agents to read page rendering results and perform click operations.
Why it matters: The original text provides the integration method for browser tools, handoff mechanisms, and WebVoyager experimental data, which can be used to determine how managed agents can cover web tasks without APIs.
LangChain BlogFeatured6767LangChain releases Managed Deep Agents v0.10, adding browser usage capabilities to let agents operate websites without APIs, and supporting asynchronous sub-agents to run long tasks in the background.
Why it matters: The original text provides specific configuration methods and handover processes for browser takeover and asynchronous sub-agents, allowing judgment on the division of labor for agents handling websites without APIs and long tasks.
The DecoderSame story · 2 outlets6565OpenAI has doubled the Work and Codex quotas for the Pro 100 and Pro 500 plans, doubled the quota limit for the Pro 200 plan and canceled the originally planned reduction, while the Plus plan remains unchanged.
MarkTechPost4747Prime Intellect released Prime Mega MoE, adding fast overlapping reverse transmission on the basis of DeepSeek's integrated Mega MoE core, making it available for training.
TechCrunch · AISame story · 2 outlets5555Meta quietly launched an AI fashion app called Style Sketch, which connects to Instagram or allows users to manually upload photos to build a virtual wardrobe. It recommends outfit combinations based on existing clothing and provides virtual try-on features matched to body and skin color. The app is currently only launched in Indonesia, Singapore, and Thailand, supporting iOS and Android platforms.
The Verge · AISame story · 2 outletsGoogle Gemini can now generate, preview, and edit PDF, Google Docs, Google Sheets, presentations, Markdown, TXT, and code files directly in the chat. This feature is now available to all Gemini users today. PDF preview/editing supports both web and mobile, while other formats are currently only available on the web version. Users can edit documents in full-screen mode, highlight text to ask questions, or let Gemini perform modifications, such as converting tables to charts.
Claude · YouTubeSame story · 2 outlets
