Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

China‑specific Apple Intelligence official website officially launches: powered by Qwen, rolling out in November

Image source for the header photo: Apple On October 31, Apple’s China website Newsroom released a press release titled “Apple Intelligence Coming to China,” outlining a specific timeline: starting November 4, the first batch of Apple Intelligence features will be available for testing through the latest iOS developer beta; and when iOS 27.3 is officially released in December, these features will roll out to Chinese users. The Apple Intelligence page on Apple’s China website has also been updated. Back in September, during the launch of the iPhone 18 and iPhone Duo, the China‑specific product page listed Apple Intelligence followed by the note “Release date subject to regulatory approval.” Even this week, with the new MacBook Pro launch, that same line remained on the China‑specific page. Now, the top of the page reads: “The first core features of Apple Intelligence will launch later this year.” Counting from its debut at WWDC in June 2024, China‑specific iPhones have waited a full two and a half years. In the early hours of March 31 this year, some China‑specific iPhone users noticed that “Siri” in Settings had changed to “Apple Intelligence & Siri”; according to reports at the time, the change was reversed within about two hours, with Apple attributing it to a software issue. On July 15, the Cyberspace Administration of China announced that seven mobile‑side generative AI services had completed registration, including “Apple Intelligence” from Apple Technology Development (Shanghai) Co., Ltd. On September 29, several China‑specific iPhone models—including the iPhone Duo and iPhone 18 Pro—completed their network access license renewals, with three new large‑model registration numbers added to the licensing documents, corresponding respectively to Apple Intelligence, Alibaba’s Tongyi Wanxiang, and Tongyi Qianwen. On October 15, Tim Cook, who has stepped down as CEO but remains Apple’s Executive Chairman, confirmed in an interview with The Paper in Shanghai: “We are very pleased to have received approval to bring the first core features of Apple Intelligence to Chinese users, including real-time translation, the removal tool in the Photos app, and writing tools, among others.” He also said he looks forward to delivering these features to users via a “beta version launching later this month.” However, the actual beta release ended up being delayed by a few days, landing on November 4—and only in the developer beta. For regular users to access them,

Timeline and related coverage

As Anthropic finally files for IPO, rising risks aren’t slowing the AI buildout – yet

The other artificial intelligence shoe dropped this week, as Anthropic filed for what surely will be the second blockbuster AI initial public offering of stock this year. Nearly 70,000 words in the filing list a ton of risks, but in short, it’s making unimaginably huge bets on buying computing power, it’s uncertain on how long […] The post As Anthropic finally files for IPO, rising risks aren’t slowing the AI buildout – yet appeared first on SiliconANGLE.

Timeline and related coverage

b12007

kv-cache : keep rotation constants on the GPU (#31073) * kv-cache : keep rotation constants on the GPU Assisted-by: Codex * tests : verify resident KV rotation buffers Assisted-by: Codex * kv-cache : document remaining multi-device rotation work Assisted-by: Codex * kv-cache : support per-layer rotation buffers Assisted-by: Codex * kv-cache : fix hybrid SWA rotation inputs and mixed placement check Assisted-by: Codex * models : use per-layer rotation in mistral-large4 and dflash * tests: drop the rotation buffer tests --------- Co-authored-by: Pascal **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b12007/llama-b12007-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](ht

Timeline and related coverage

b12005

opencl: add initial q2_0 support (#31340) * opencl: add q2_0 AOS mv Co-authored-by: Shaofei Qi Assisted-by: Codex * opencl: add q2_0 SOA mv and tiled mm Co-authored-by: Shaofei Qi Assisted-by: Codex * opencl: add q2_0 noshuffle gemm/gemv for adreno Assisted-by: Codex * opencl: add q2_0 dp4a for adreno * opencl: optimize q2_0 adreno decode and narrow batches --------- Co-authored-by: Li He **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b12005/llama-b12005-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b12005/llama-b12005-bin-macos-x64.tar.g

Timeline and related coverage

Claude’s multi-platform app has received a major upgrade, with the iPhone version now officially supporting CarPlay voice interaction.

Recently, Anthropic announced on X that its cross-platform mobile app has undergone a comprehensive and in-depth upgrade. Most notably, the iPhone version of Claude’s voice mode has been officially integrated into Apple’s CarPlay system, allowing users to engage in seamless voice-based Q&A with Claude directly through their car’s infotainment system, delivering a brand-new intelligent human–vehicle interaction experience. Beyond in-car scenarios, this update also extends to interface and other multi-device features.

Claude多端应用迎来重磅升级,iPhone版正式支持CarPlay语音交互
Timeline and related coverage

Claude gets into the car: With CarPlay integration, you can even ask AI questions while driving.

On October 29, Anthropic announced an update to the Claude mobile app, supporting iPhone, iPad, Android, Chromebook, as well as web and desktop platforms. A key highlight is the integration of the iPhone version’s voice mode with Apple CarPlay, allowing drivers to ask AI questions hands-free—perfect for checking routes, finding information, and organizing thoughts during commutes. The in-car use case is seen as the most imaginative aspect of this upgrade.

image.png
Timeline and related coverage

Morning Briefing | Apple executive responds as iPad mini remains at 60Hz / Global smartphone shipments fell 20% in Q3, hitting a new low / "Shadow Blade Zero" tops Steam's global best-sellers chart

· Wang Chuanfu reveals BYD’s workforce has surpassed 1.02 million, with R&D engineers numbering 120,000 · U.S. computer science enrollments decline again, down 18.4% over the past two years · South Korea plans to extend its ban on AI glasses during exams to three years and will deploy dedicated personnel for identification #Follow iFanr’s official WeChat account: iFanr (WeChat ID: ifanr) for more exciting content delivered straight to you.

Timeline and related coverage

b11985

metal : faster MXFP4 mat-vec, full threadgroups for unary/binary/GLU (#31075) * metal : faster MXFP4 mat-vec kernel Decode the fp4 values with the fp16 bit trick instead of a threadgroup lookup table, and load the packed nibbles as aligned 32-bit words. The kernel no longer uses threadgroup memory. * metal : launch contiguous unary and binary ops with full threadgroups The contiguous unary path and the row-broadcast binary path used one thread per threadgroup. Use up to 256 threads per threadgroup and bound-check the tail. * metal : split GLU rows across threadgroups A GLU row ran on one threadgroup, so single-row decode used a single threadgroup for the whole op. Split each row over threadgroups of up to 256 threads. **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon

Timeline and related coverage

b11982

hexagon: support for UPSCALE op (#31331) * hex-upscale: basic upscale ops (not fully vectorized) * hex-upscale: optimize nearest 2x, 4x and bilinear 4x * hex-upscale: add support for non-contig rows and dims collapsing * hex-upscale: use vgather instead of scalar staging * hex-upscale: move all kernel select logic to the host, review comments and cleanups * tests: add permuted UPSCALE test to cover qwen3-vl image->text scenarios **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11982/llama-b11982-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/downloa

Timeline and related coverage

b11981

hexagon: improve DDR bw utilization in transposed CONCAT (#31277) * hex-concat: double row block size for transposed CONCAT * hex-concat: double buffer scratchpads * hex-concat: chunked src1 rows in transposed concat * hex-concat: pad src1 VTCM stride to spread gather lanes * hex-vgather: apply similar vgather opts to SSM_CONV and GDN --------- Co-authored-by: Max Krasnyansky **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11981/llama-b11981-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11981/llama-b11981-bin-macos-x64.tar.gz) - [iOS XCF

Timeline and related coverage

Meta quietly launches the AI pet diary app Waddle World

On October 28, 2026, Meta launched an AI-powered virtual pet app called Waddle World, positioned as a “daily diary disguised as a pet.” Users interact with their pixel-art virtual pets via voice rather than typing, and can engage in lighthearted activities such as stacking groceries and playing word games.

新闻源配图 · The Verge · AI

Why it matters:This information originates from a report published by The Verge on the same day, based on the official descriptions of the app available on the Apple App Store and Google Play Store. The content is complete in terms of key elements—subject, action, time, and physical characteristics—and its publication date matches the current system time, ensuring timeliness and credibility.

Timeline and related coverage

b11954

tests : adjust constants to reduce irrelevant warnings (#31254) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11954/llama-b11954-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11954/llama-b11954-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11954/llama-b11954-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11954/llama-b11954-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b1

Timeline and related coverage

b11953

cuda : add top-k MoE fusion for 384 experts (#31288) Assisted-by: Codex **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11953/llama-b11953-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11953/llama-b11953-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11953/llama-b11953-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11953/llama-b11953-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/dow

Timeline and related coverage

b11951

chat : stop qwen3-coder reasoning and content before template separators (#30557) * chat : strip trailing whitespace from parsed reasoning and content Assisted-by: Claude Fable 5.1 * chat : stop qwen3-coder reasoning and content before template separators Assisted-by: Claude * tests : remove the Qwen parser case from the chat session tests Assisted-by: Claude * common : handle partial prefix until match at end of generation --------- Co-authored-by: Zeshawn Zahid Co-authored-by: Alde Rojas **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11951/llama-b11951-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel

Timeline and related coverage