NVIDIA BlogThe NVIDIA Jetson Orin NX module is used for inference on the pilot line of AIRECO's recycling plant, driving a robotic arm to pick up target objects from a conveyor belt with an accuracy rate of 95%; plans are to increase the annual processing capacity from 18,000 tons to 48,000 tons next year, reducing operating costs by 85%.
Microsoft announces the official launch of its AI super-resolution technology Automatic Super Resolution, supporting Asus ROG Xbox Ally X handheld and some AMD Ryzen platform Windows Copilot+ PCs, with a maximum frame rate of 120 FPS. The technology can enhance game screen resolution in handheld mode, adds support for 1080p games, and supports running simultaneously with Windows Auto HDR.
AMD and Spectro Cloud launch Instinct Coder 355, supporting local deployment of GLM-5.3 inference with 98% performance. The system is equipped with 8 AMD Instinct MI355X GPUs and 2 x 64-core EPYC 9575F CPUs, featuring 3TB DDR5-6400 and 63.36TB SSD, with the Dell version supporting up to 85 developers per node.
MooreThread announced that its MTT S5000 AI computing card and MUSA software stack have been adapted to the MiniMax H3-Image image generation model. This model can generate high-resolution images of 4096×4096 pixels within 10 seconds, achieving speeds five times faster than mainstream market offerings. MooreThread has optimized the entire chain based on MTT S5000, supporting text-to-image generation, image reference, and editing capabilities.
AWS Machine Learning BlogMoonshot AI's Kimi K3.1 Pro Highspeed has been released on Amazon Bedrock, the model being a post-training version of the Kimi K3.
Comfy BlogSeedream 5.5 is provided through partner nodes in ComfyUI, a multimodal model that generates content from text and edits from images. This version advances in transparent RGBA generation, precise local multi-round editing, poster and chart generation with readable text and structured layouts, more realistic portraits, and lighting control. In ComfyUI, all features are integrated into a single node, allowing for sequential editing processes and direct transparent output to synthesis nodes.
Microsoft ResearchMicrosoft researchers discover 'Lazarus' technique can reprogram AI model dormant neurons to plant backdoors, enabling normal AI agents to execute malicious commands under trigger conditions. The technique can plant backdoors without retraining and is difficult to detect through conventional tests, posing a particularly severe threat to AI agents with tool or terminal access.
Claris is testing the Harbor app, which can turn a single or multiple Apple Silicon Macs into a local AI server to run larger AI models without manual configuration. The app is currently in the early experience phase and has not been made available to the public yet.
KORA Systems releases the new operating system KORA OS, claiming it is the first 'bottom-up' system designed specifically for AI PCs, integrating agent workflows, contextual memory, and local AI reasoning at its core.
OpenRouter BlogThe GitHub MCP server added 19,436 input tokens for each request of Claude Sonnet 5.5, far exceeding the CLI (1,421) and skills (3,315).
OpenRouter BlogTesting on OpenRouter, fp8, fp4, and fp16 showed no precision differences in 240 scoring tasks, with all six endpoint scores ranging from 98.3% to 99.2%. Lower precision is generally cheaper, but there were 7 models where fp8 was more expensive than fp16, and 7 models where 4-bit was more expensive than 8-bit. The fp4 endpoint performed the worst among 40 workflows without multiplication tasks, but the difference might be due to chance.
Gigabyte introduces the AI TOP AERO mini PC, equipped with the NVIDIA RTX Spark N1X super chip, integrating a 20-core Grace CPU and Blackwell architecture GPU, with optional up to 128GB LPDDR5X unified memory and 4TB PCIe Gen 5 SSD. It features USB Type-C, 10GbE Ethernet, and Wi-Fi 7 interfaces, housed in a compact one-liter chassis design.
AWS Machine Learning BlogDevelopers used LangGraph and Amazon Bedrock AgentCore Web Search to build a daily dependency monitor, watching daily changes in open-source software packages and container base images. The monitor checks daily new impacts on team releases, searches using Anthropic Claude models, and uses the Web Search API to get current results.
Google launches VLA robot model Gemini Robotics 2.5, achieving leaps in dexterity, reliability, and safety, with a success rate of 96% on the Mercedes production line. The model has tested 200 tasks, with a success rate 30%-100% higher than the baseline, a speed increase of 20%-30%, and is compatible with robots like Apptronik Apollo 2, adopting ASIMOV-Jailbreak benchmark to enhance jailbreak resistance to 99.5%.
Google has released the Gemini 4 Carbon model, which is its most powerful reasoning model to date. The model has a unified context window of 1,048,576 tokens, with a maximum standard synchronous request output of 65,536 tokens, expandable to 1,048,576 tokens via the Long Decode mechanism.
Anthropic introduces Claude Whiteboards, Claude Sheets, and Claude Data, moving AI assistants from chat Q&A to team collaboration workstations
PyTorch BlogPyTorch models run on WebGPU in browsers via ExecuTorch without converting framework formats. Export as ExecuTorch .pte files and run on any WebGPU browser using the new ExecuTorch WebGPU backend.
AWS Machine Learning Blog4040Amazon Bedrock AgentCore Evaluations provides three evaluators—built-in, DeepEval, and AutoEval—to assess agents. The built-in evaluator is developed and tested by AWS, covering common dimensions; DeepEval and AutoEval are third-party open-source libraries, with AgentCore hosting them in production environments. The evaluation process requires first defining business success criteria, breaking them down into observable behaviors, and then mapping them to the appropriate evaluator level.
OpenAI Alignment BlogGPT-6 Astra shows a decrease in CoT monitoring ability in non-adversarial settings, an increase in CoT evasion ability in adversarial tests, and no significant improvement in CoT capability. During pre-training, CoT capability does not increase with the decrease of the base model's perplexity, and the architecture has a limited impact. Reinforcement learning enhances CoT control ability, and model-generated data fine-tuning can transfer CoT control ability. Model perplexity is positively correlated with CoT capability and control ability.
LangChain Deep Agents releases3535LangChain Deep Agents deepagents-code v0.2.7 released, adding shell completion scripts, fixes Windows editor launcher detection, `dcode completion` output, and Bash completion issues, improves Git attribution performance and adds a disable option.
The DecoderNvidia's GR00T 2 AI model learns new tasks for robots by analyzing human video actions without remote control, trained on 400,000 hours of video data, with lab test success rates of 10-80%, and combined with GPT-6 Astra achieves 97% success rate for toy model assembly, but tactile processing and language model latency remain challenges.
Google AI BlogGemini Intelligence is now available on more devices, including the Motorola Signature 27 and Pixel 10. New features will expand capabilities on these devices, helping users complete more tasks. Create My Widget is launching soon, allowing users to create personalized widgets with prompts; task automation will expand to the UK, India, and Japan.
Latent SpaceNaive-N0.5-Flash is an open-source coding model developed by NaiveAI, achieving a high throughput of 2,100 tokens per second by improving Xiaomi's MiMo-V2.5 and integrating the NaiveRT engine.
AWS Machine Learning Blog
Codex releases00GitHub Copilot 0.167.1 is released, providing features such as code review, vulnerability detection, and secret scanning.
Google Developers BlogGoogle AI Edge Gallery has launched on-device automatic speech recognition (ASR) and text-to-speech (TTS) capabilities, enabling fully offline operation of multilingual ASR and natural TTS models on phones and wearable devices.
Meta EngineeringMeta open-sources Triton Low-Level Extensions (TLX) compiler, enhancing GPU performance by introducing explicit warp-group and cluster-level orchestration, supports RecSys training/inference and LLM pre-training, achieving end-to-end performance improvements over standalone libraries on NVIDIA and AMD GPUs.
Meta EngineeringMeta has launched the Your Algorithm tool, allowing users to edit interest lists ("Interested In"/"Not Interested In") in natural language to adjust Feed and Reels recommendations in real-time. The system uses LLM to process user preferences and integrate user behavior, achieving responsive adjustments through retrieval, reordering, and pushing optimization stages. New interest content will appear quickly but will not dominate recommendations.
LangChain BlogWorkday has utilized LangGraph for agent orchestration and LangSmith for tracking and evaluation, reducing prototype development time from 5 months to within 1 day with a 4-person team by 2025 and supporting production environment deployment. Its Python SDK 'Workday LangChain' encapsulates LangChain core components and routes through internal services (including LLM gateway, RAG retrieval) while meeting enterprise-level security and compliance requirements.
PyTorch BlogExecuTorch WebGPU backend allows PyTorch models to run in WebGPU browsers without converting framework formats, supporting multiple models such as Llama 3.2, Qwen3, and Gemma 3.
AWS Machine Learning BlogTCS has built an Agentic test orchestrator for BFSI clients, transforming requirements into executable controlled tests through Amazon Bedrock AgentCore, breaking down tests into dedicated tasks and executing them using their respective agents, tools, and validation rules, while meeting project-specific context, API, and browser UI testing, tenant-level security isolation, and repeatable test path requirements, significantly improving test throughput and allowing engineers to focus on risk analysis and exploratory testing.
OpenRouter BlogOpenRouter tested three methods for obtaining structured outputs from LLMs, validated on nine models supporting structured outputs and only tool-supported GLM 5.3 FlashX, where 209/216 responses matched strict JSON Schema.
OpenRouter BlogOpen-source multimodal model Claude Code v2.1.159 released, supporting multilingual and code generation.
n8n BlogSame story · 5 outlets6262Google Gemini 4 Carbon has officially launched on n8n, being Google's strongest model, scoring 53.9% on AutomationBench. It significantly improves in security (SEC-bench Pro 88.2%, ExploitGym 54.8%, CWE-bench 71.0%), long task processing (internal testing lasting several days), and reasoning, and has zero vulnerability to Gray Swan prompt injection attacks (0.0%).
Codex releases00GitHub Copilot released version 0.169.0-alpha.9, without mentioning specific update content.
Factory NewsSecurity Factory connects the signal, classification, and remediation three stages to help engineering teams transform security discoveries into test fixes. It integrates existing tools and contextual information, automatically coordinates investigation tasks, generates a risk-prioritized remediation list based on validation, and simplifies the entire process from discovery to remediation.
Amp News6565Amp launches Boxes, a permanently running, small, and affordable computer where users can deploy software built with Amp, with the smallest model priced at $18 per month, supporting the operation of any application and equipped with an AI agent to manage deployed applications and production data. Users can create, deploy applications, and control the power on/off of Boxes via the agent as needed, supporting minute-by-minute billing and default continuous operation.
Google DeepMindFeaturedSame story · 5 outlets8787Google DeepMind has opened Gemini 4 Carbon to developers, enterprises, scientists, and consumers, claiming it has reached cutting-edge levels in network security capabilities and long-range intelligent agent workflows.
Why it matters: The official provides the positioning division of roles between Carbon and Argon, multiple benchmark rankings, and cost comparisons per task to help determine when to use a heavier model.
NVIDIA Technical Blog6565The NVIDIA Alpamayo 2 Edge edge inference model can run within automotive computing and latency constraints. The model achieves 70.5 on the MMLU benchmark, supports an 8k context window, speeds up 3.5x, and costs $3/ million tokens with closed-source deployment.
