Skip to content
Live
#Engineering
Oct 29Thu
  1. The Verge · AI50

    The New York Times cooking app is testing the AI recipe modification feature

    The New York Times Cooking app has launched new AI features that can recommend ingredient and kitchen tool alternatives, adjust quantities, and modify recipe steps on demand. Currently available to iOS subscribers, it will gradually be rolled out to a wider user base. Some users are concerned about the AI's impact on user experience and the environment, asking if the feature can be turned off. The editor responded that it is 'not possible' at the moment but described it as 'fully optional.' This is not the first time the app has used AI, having previously been used for unit conversion and quantity adjustment.

  2. Hugging Face Blog

    Project ProtoStar project release: Promoting the AI compatibility of publicly funded scientific research

    Project ProtoStar is announced, aiming to unlock the potential of publicly funded scientific research by making datasets, models, and code more accessible and reusable. The project brings together universities, research institutions, industry, and the open science community, supporting the release of scientific achievements in AI-compatible, machine-actionable formats, and demonstrating its value through a series of public challenges. The first challenge, hosted by Carnegie Mellon University, will launch on November 2.

  5. The Verge · AI78

    Anthropic's S-1 filing impacts its IPO, with Wall Street being asked to believe in AGI.

    Anthropic has filed an S-1 registration with the Nasdaq, aiming to list as soon as next month with a valuation range of 1.5 trillion to 2 trillion U.S. dollars. Analyst Joey Brookhart from SemiAnalysis dissected the filing on the Decoder podcast, noting the company's ARR has exceeded 70 billion U.S. dollars, with an approximate 85% reasoning gross margin, but revenue is highly concentrated in programming scenarios, with the top 1% of customers contributing most of the spending.

    Same story — showing the featured version《Anthropic files S-1 with SEC, plans NASDAQ IPO》

  6. Hugging Face Blog85

    Mistral Large 4 (Le Chonk) released on Hugging Face

    Mistral releases multimodal MoE model Mistral Large 4 (Le Chonk), with approximately 1T total parameters, 49B activation parameters, a context window of 1M tokens, offering two official versions FP8 and NVFP4, as well as GGUF made by Unsloth and llama.cpp teams.

    Same story · 2 outlets

    Why it matters: The original text provides specific scores for the 1T total parameter, 49B activation multimodal MoE model on cybersecurity and coding benchmarks, along with measured throughput for local and endpoint deployment.

  7. Hugging Face Blog

    Hugging Face Granite Switch 4.1.1 released: software-based model update

    Hugging Face releases Granite Switch 4.1.1, an incremental update based on 4.1, maintaining full backward compatibility. New features include multilingual speech-to-text models (supporting 10 languages with native punctuation) and a decision adapter for improved performance on the Jev decision index test set. The team achieves rapid deployment by updating only the adapter without recalculating the base model, and future support for third-party custom adapters is planned.

  8. ITHome · AI58

    Microsoft has released the largest DirectX ray tracing upgrade since 2018: AMD RX 9070 example performance improvement of about 66 times

    Microsoft has released the official version of DirectX 12 Agility SDK 1.620 and the preview version 1.722, bringing updates such as cluster geometry and PTLAS to DXR to reduce ray tracing computational overhead. On the AMD RX 9070, cluster geometry boosts performance from 1.2 FPS to 79 FPS, an increase of about 66 times, with memory usage reduced by nearly 90%. New features require support from graphics card manufacturers, currently covering NVIDIA RTX, AMD RX 9000 series, and Intel Arc graphics cards.

  9. 36Kr · AI

    Zhiyuan Innovation releases the GO-3 Preview, WITA-OMNI, AGILE model combination to accelerate the deployment of the entire physical AI chain

    Zhiyuan Innovation launches GO-3 Preview (tops RoboDojo and other evaluations), WITA-OMNI (wins first place in four rankings), and AGILE general sports intelligence algorithm framework, respectively tackling tasks, interaction, and sports abilities; Lingqu OS 1.0 and AHA3 hardware architecture integrate software and hardware ports; Worken measures intelligent work hours, aiming to convert 20000 offline robots into actual productivity.

  10. Mistral AI82

    Mistral releases open-source model Mistral Large 4 (Le Chonk) with trillion parameters

    Mistral releases open-weight version of its flagship trillion-parameter native multimodal model Mistral Large 4 (Le Chonk), which has been in public preview for three weeks.

    Same story · 2 outlets

    Why it matters: The official provides data on the encoding, security, and inference efficiency of the trillion-parameter open-source model, as well as weight downloads and self-hosting paths.

  11. TechCrunch · AI65

    Amazon updates Alexa to help users manage shopping lists, subscriptions, and update delivery instructions

    Amazon has added over 100 new shopping-related features for Alexa before the holidays, including subscription and list management, updating delivery instructions, resolving payment issues, and answering questions about Prime subscriptions. This update allows users to use Alexa to create, register, view list contents, and add or edit items, manage subscriptions, such as changing the delivery frequency of an item from once a month to once every three months, and edit delivery instructions, such as changing an address or adding pickup details.

  13. Hacker News · AI hot posts

    DAYJOB: Tax released, evaluating the performance of AI agents in 50 professional tax tasks.

    DAYJOB: Tax evaluates the performance of AI agents in 50 professional tasks across 8 tax areas, requiring agents to complete tax filing, compliance, and advisory work. Claude Sonnet 5.5 leads the rankings with a pass rate of 10.8%, while Claude Opus 5.5 has a pass rate of 9.2%. The benchmark includes 1,789 workspace files, covering various tax scenarios and complex materials, created and reviewed by tax experts.

  14. QbitAI72

    Tenstorrent releases Tianhe 300 and Tongyang TY1200, on-site actual test targets NVIDIA H200 and Thor T5000.

    Tenstorrent releases cloud-based general-purpose GPU Tianhe 300 and edge product Tongyang TY1200 based on the Tianshu architecture, and compares them with NVIDIA H200, Jetson Thor T5000 on-site with random questions.

    Why it matters: On-site random tests, dual-track comparison of cloud and edge against NVIDIA, and providing reproducible test numbers and software stack implementation methods.

  15. QbitAI72

    Genspark releases Media Studio video Agent workbench

    Genspark releases video Agent workbench Media Studio, integrating over 70 model versions from more than 20 manufacturers, where large language models act as directors responsible for writing scripts and scheduling video, image, and audio models.

    Why it matters: The author conducted a complete test of the entire process from short drama to knowledge explanation video, allowing readers to judge the completeness and shortcomings of multi-model arrangement in real-world creation based on this.

  16. ITHome · AI42

    XHEART NewCore Journey releases the intelligent driving large model chip X7 Ultra: 700 TOPS large model computing power, first in China to mass-produce Armv9 automotive-grade CPUs.

    XHEART NewCore Journey releases the X7 Ultra large model chip, achieving 700 TOPS of computing power, with mass-produced models achieving 99.99% alignment, a 80% reduction in memory bandwidth consumption, and a 100% improvement in energy efficiency. The chip incorporates a high-precision operator acceleration engine, natively supporting FP16/FP32 computing, and is the first in China to mass-produce Armv9 automotive-grade CPUs, offering a 20% performance boost and a 10% improvement in energy efficiency.

    Same story · 2 outlets
  17. MarkTechPost

    AssemblyAI releases AssemblyAI-TTS-1: The first text-to-speech model, API price is 15 USD per 1M characters

    AssemblyAI has released its first text-to-speech model AssemblyAI-TTS-1, targeting voice agents in real-time phone calls, priced at $15 per 1M characters, with no annual contract and support for unlimited concurrency. The model achieved an entity accuracy rate of 93.7% in AssemblyAI's tests, comparable to ElevenLabs v4, but priced only one-fifth of it, with a median first byte time of 47 milliseconds, but the median streaming latency in Cekura's independent benchmark tests was 159 milliseconds.

  18. MarkTechPost

    Infini-AI-Lab releases open-source AI agent adapter framework Hermit

    Infini-AI-Lab releases open-source framework Hermit v0.1.0, with adapters that can fine-tune Codex or Claude Code agents without altering the model or framework. Using GLM 5.3 Flash on OfficeQA, CL-Bench, and ServeLearnBench, results with adapters are better than baseline results without adapters. The MWM adapter achieves 0.795 accuracy in database exploration on CL-Bench.

  19. QbitAI

    Zhuji Dynamics releases the FluxVLA Engine V2 technical report, enhancing the R&D efficiency of physical intelligence.

    Zhuojì Power updates the second technical report of the open-source engineering base FluxVLA Engine. V2 supports 14 strategies for LIBERO evaluation on the simulation side, increases model path frequency by 2.31–9.64 times on the inference end, and verifies deployment links on real hardware platforms such as ALOHA and dual-arm Franka. Online reinforcement learning boosts SmolVLA's success rates in LIBERO-10 and RoboTwin Clean tasks by 10.0% and 15.4%, respectively.

  20. OpenRouter Blog

    OpenRouter has refactored its supplier routing algorithm to improve reasoning performance and speed.

    OpenRouter has released a new supplier routing algorithm that selects the best cost-performance solution for each request by evaluating price, latency, throughput, and error rate in real time, significantly improving reasoning speed (TTFT improved by 16.9%), throughput (improved by 16.3%), and reliability (error rate reduced by 67%), with only a slight increase in cost of 6.0%. This algorithm rewards suppliers that provide high-quality reasoning without requiring users to modify their code, ensuring a continuously optimized experience.

  21. OpenRouter Blog

    How to Use LLM to Extract Tables from PDFs

    The Anthropic/claude-haiku-5.5 model paired with a strict JSON schema can extract tables from PDFs and return typed rows. Tested five PDFs (including scans) successfully extracted all 644 cells; failure cases were due to PDF processing paths, not model capability. Note empty results and output length limits.

  23. 36Kr · AI

    NVIDIA has launched a local AI speech model, with new models and agent tools added to RTX Spark.

    The Nemotron 3.5 Transcribe open-weight speech model now supports RTX Spark, enabling speech transcription, speaker recognition, and translation, supporting 40 languages. The Liquid AI LFM3 series models have also been supported and can be run locally and fine-tuned. The RTX Spark laptop starts at $2599, with the highest configuration of 128GB unified memory, and FP4 AI computing power up to 1 petaflop.

  24. TechCrunch · AI

    Google is expanding its AI search advertising business to the global market

    Google has expanded AI-powered search ads to international markets, trying to open up new revenue streams as core search ad growth slows. Q3 revenue grew by 25% year-on-year to $127.6 billion, but search ad growth continued to decelerate, and Alphabet's stock price fell 8%. Philipp Schindler said in the earnings call that high-quality ads have been extended to more users and queries in global AI overviews and AI patterns, with a click-through rate comparable to natural search results.

  26. Sierra Blog65

    Sierra introduces Compliance Console to help businesses monitor the compliance of long-term AI agents

    Sierra introduces Compliance Console, enabling businesses to set compliance rules for long-term AI agents through the Horizon platform, covering dimensions such as channels, use cases, and regions. Users can adjust control items like disclosure requirements and frequency limits to comply with local regulations. This tool supports businesses in expanding their agents to new scenarios and countries while adhering to the rules of the customer's location.

  28. Databricks Blog

    Databricks launches voice agents, delivering real customer value through context.

    Databricks releases voice intelligence agents, which adjust responses in conversations based on real-time contextual decisions, enabling language switching, transferring to a human, or providing personalized services. These agents need to process customer requests within 200 milliseconds and rely on the context, ontology, and governance capabilities provided by Databricks. They integrate customer, order, product, and policy data, and follow the access rules of the enterprise data catalog.

  31. Google Developers Blog58

    How Agents CLI optimizes Agent quality through Hill-Climbing

    Agents CLI v1.8.0 achieves Hill-Climbing optimization of Agent quality through commands like eval generate, eval grade, etc., using the custom_response_quality metric to rate 4 cases from 1 to 5, finding that changing the system prompt to a chef role reduced the average score from 4.0 to 2.75. This tool supports model backend experiments and automatic alerts, improving Agent performance through iterative cycles.

  32. Baseten Blog

    Optimizing Baseten for Qwen3-ASR 1.7B to achieve low-cost high-throughput batch speech transcription

    Baseten achieved the lowest cost per minute for batch speech transcription by optimizing Qwen3-ASR 1.7B (175B parameters, 5.4% WER) on Baseten. Its vLLM stack supports 7 times the throughput of standard vLLM on RTX PRO 6000, with a maximum real-time factor of 3,128, enhanced by native audio decoding, FP8 KV cache quantization, and reduced host-device synchronization.

  33. LangChain Blog

    How to use crons to build proactive agents with LangChain

    LangChain found that nearly 70% of Managed Deep Agents run from scheduled tasks. Crons enables agents to have proactivity, personalization, and ambient characteristics by implementing mechanisms such as memory ('dreaming'), proactivity (exploring new actions), intelligent filtering (monitoring event streams), and reminders (scheduled tasks).

  34. The Decoder80

    Google announced the integration of third-party AI agents into Android Studio at the Android Dev Summit and released a new Gemini API

    Google announced Android Studio Rabbit 2 at the Android Dev Summit, supporting more than 40 coding agents, including Claude, Codex, and Google Antigravity, to run within the IDE through the open-sourced Agent Client Protocol.

    Why it matters: Android Studio opens third-party intelligent agent access and provides corresponding changes at the CLI and MCP levels, which can be used to judge the changes in mobile development workflows.

  37. TechCrunch · AI65

    Chrome browser introduces an AI-driven Autofill feature that can extract information from Gmail to automatically fill out forms

    Chrome browser updates its Autofill feature, using AI to extract information from users' Gmail to automatically fill out forms, including flight confirmation numbers, member IDs, tracking information, order numbers, addresses, phone numbers, or credit card numbers, etc. This feature expands Google's Personal Intelligence, and users can choose whether to enable it. Sensitive data will be encrypted. The feature has been launched in the US desktop version of Chrome and Chrome versions for iOS and Android.

    Same story · 2 outlets
  38. The Verge · AI

    Chrome's enhanced auto-fill feature can now extract information from Gmail to help fill out forms

    Chrome's enhanced auto-fill feature can now extract information from Gmail to help fill out forms, such as shipping numbers, flight reservation numbers, or hotel reservation numbers, while keeping sensitive data encrypted. This feature is implemented through Gemini with Personal Intelligence and is currently rolling out in the U.S., supporting desktop, Android, and iOS versions of the Chrome browser. Users can manage access permissions or completely turn off the feature in settings.

    Same story · 2 outlets
  39. Artificial Analysis Articles72

    Artificial Analysis Reviews GitHub Copilot Fusion: Performance slightly lower than Opus 5.5, single-task cost halved

    Artificial Analysis tested GitHub Copilot CLI's Copilot Fusion highest settings in Coding Agent Index v2.0, with a single task cost of 5.59 USD, while Claude Code with Opus 5.5 (max) costs 11.42 USD, only 2 points lower in overall score.

    Why it matters: Artificial Analysis used Coding Agent Index to test Copilot Fusion, providing a cost and token comparison with Claude Code at the same performance level.