MarkTechPost6060DeepSeek released dsh-superagent, an MIT-licensed proactive coding agent prototype, v0.1.0, which routes tasks to isolated worker sessions for execution, simulating test p95 task completion time of 15.7 minutes.
MarkTechPost5858Nace AI open-sources its most powerful decision model, Drex Max v1, which can run locally on Apple Silicon hardware. The model includes bf16 weights, Q8_0 GGUF format, affine Q4 checkpoints, and llama.cpp/Ollama branches, scoring 67.52 on the public decision index 0.4, surpassing Drex v2.0 (61.25), and achieving a 97.3 skill score on InfographicVQA.
Qualcomm released the Adreno GPU Windows driver version 2026.10.1, improving the performance of the Snapdragon X and Snapdragon X2 processors in games such as 'Naraka: Bladepoint,' 'Red Dead Redemption 2,' and 'Assassin's Creed.'
OpenRouter Blog6060The AI Code Reviewer built using the OpenRouter Responses API can check for errors and security issues in Git code changes before submission, with an average review time of 1-4 minutes and a cost between $0.015 and $0.045.
NVIDIA demonstrates next-generation co-packaged optics technology OOI, integrating the optical engine onto the GPU silicon interposer, bringing the photoelectric conversion stage closer to the chip core, prototype samples are ready. This solution replaces the current organic packaging substrate with a larger area interposer, currently in the research stage, and has no commercialization timeline.
Megvii open sourced the SenseNova Loop-9B-Preview model, which uses the Looped Transformer architecture and matches the performance of 27B models with just 9B parameters, outperforming other models including surpassing Qwen3.5-27B in code generation, office scenarios, and intelligent agent tasks
GPT-5.6 Sol completed a Factorio run in peaceful mode (no biters) in Codex with low thinking levels, taking 13.5 hours of game time. The run used a game manager, allowing Sol/Codex to interact with the game through a small library (Python instead of Bash this time), and the model programmed the game according to GUI rules. Observing the run showed that Sol exceeded expectations in some aspects but also exposed clear weaknesses.
Databricks Blog5858Databricks Lakebase reduces the creation time of each PR branch to seconds with its copy-on-write database branching feature, supporting up to 100 concurrent branches, with each branch having independent read/write, compatible with standard Postgres, and supports large JSONB/LOB columns, meeting Snorkel AI's needs for fast, secure, and low-cost testing of production data in CI/CD processes.
Browser Use6666Browser Use launches BU Ultrafast Browser Agent, reducing task processing time to 22 seconds and cost to 0.0064 USD per task, 3 times faster and cheaper than the default GPT-6 Luna model. The model is now freely available in all Browser Use Cloud plans in the United States, previously in a candidate status.
Latent Space4242Arca stated at the AI Engineers' New York Summit that its agents primarily handle background work, but human advisors retain final decision-making authority. Agents only analyze and propose solutions, with transactions executed by humans. Agent tasks can last up to ten days, and the company has developed an internal memory system to track task progress and responsibility allocation. They use models from Anthropic and OpenAI, automatically selecting models based on internal benchmarks via a custom harness. In terms of security, agent operations are strictly limited, involving no client risk, and they use sandbox computing and additional measures to handle sensitive data.
Arrowmax launched the AI Electric Screwdriver S1 on Kickstarter, claiming it to be the world's first, with a starting price of $80 and an estimated shipping date of April 2027. The S1 uses a camera and AI model to identify screw types and sizes, and automatically selects bits and adjusts torque, with all training samples collected in-house. The single unit starts at $80, and the kit has an early bird price starting at $129.
New Chip Voyage releases the X7 Ultra chip, describing its performance for the first time using 'large model computing power'. The NPU computing power reaches 700 TOPS with a utilization of 98.7%, the quantized model output similarity to the FP32 benchmark model exceeds 99.99%, memory bandwidth consumption is reduced by 80%, and energy efficiency is improved by 100%. This chip is designed specifically for mass production of intelligent driving large models, featuring a super-large core NPU architecture and a dedicated computing engine, paired with AI toolchain optimization to support kilowire-level LiDAR processing and camera latency jitter optimization. It is equipped with an Arm v9 automotive-grade CPU and meets ASIL-B/ASIL-D safety standards.
OpenCode launches Cheepseek Phase 4, with monthly $10 subscribers eligible for a $100 value DeepSeek V4.1 Flash model credit. This model has 552B parameters, supports text and image input, function calls, and agent workflows, with an approximate context window of 1 million tokens.
Hugging Face Daily Papers5555VibeGame Framework constructs complete 2D games from natural language prompts using an AI-native engine and eight specialized adversarial intelligence teams, supporting interactive validation and reusable self-evolving components. On the GameCraft-Bench subset, VibeGame's P2G (Prompt-to-Game) version surpasses multiple AI agent solutions like GPT-6.1-Sol with the same underlying model.
Microsoft released Windows App SDK 2.6.4 to improve the Win11 native application framework WinUI experience, avoid memory usage issues of 'webshell' applications, enhance application stability and compatibility, and strengthen AI capabilities. This version simplifies Language Model API access, upgrades ONNX Runtime to 1.27.1, fixes text control crashes, and includes reliability and diagnostic improvements covering WinUI 3, application lifecycle, etc.
OpenRouter Blog4242Open weight models on OpenRouter accounted for 72% of the top 50% model tokens by the end of September, an increase of 455% from October 2025; while proprietary models also saw usage grow nearly tenfold, the cost per million tokens input for open models is only about a quarter of that of proprietary models; due to fierce competition, open models like DeepSeek V4.1 Flash and GLM-5.3 Flash are priced lower, but the top proprietary model, Claude Opus 5.5, still leads by 11 points in AI analysis intelligence index.
NVIDIA Technical Blog6565NVIDIA Dynamo Distributed Inference Framework needs validation through GPU drivers, resource allocation, scheduling, networking, and monitoring to successfully deploy on Kubernetes. Deployment is just the first step; AI application teams need a repeatable path from model configuration to inference endpoints.
NVIDIA Technical Blog6565Developers are building next-generation intelligent agents at the edge for monitoring driver attention and cabin conditions, helping robots understand environmental planning tasks, and coordinating machine clusters on production lines. These physical AI applications require models for perception, reasoning, and action that run within strict memory, power consumption, and latency budgets.
Baseten Blog6262Baseten reduces the incremental weight sync time for the ~770 GB GLM-5.3 model from 36 seconds to 15 seconds, suspension time from 6 seconds to 0.8 seconds, and throughput increases 3.2 times.
Hugging Face Blog4040MonsterHouse is an interactive projection built with NVIDIA Nemotron Voice Agent, NVIDIA Nemotron ASR, NVIDIA Nemotron 3.5 Bolt E4B 4GB, Magpie TTS Zeroshot, and Audio2Face-3D, which can communicate with visitors through voice and imitate the voice of a grumpy old pirate.
TechCrunch · AI3838Zest launches an AI restaurant manager that recommends and books restaurants via SMS, supporting linking services for OpenTable, Resy, and SevenRooms. It can be integrated into group chats and supports voting features. Users can get recommendations based on cuisine, occasion, etc., and inquire about restaurant details.
AWS Machine Learning Blog3838
O'Reilly Radar3434O'Reilly Media CTO Andrew Odewahn's research agent maps organizational information systems, connects MCPs, and empowers skill documents, combining Expert Intelligence tiered analysis.
NVIDIA Technical Blog6565Developers use NVIDIA NemoClaw to build autonomous agents, a framework that includes models, executors, and domain-specific tools, and supports secure deployment. NemoClaw provides policy enforcement runtime, optimization, and observability tools to ensure the reliability and security of agents when accessing real enterprise systems and data.
NVIDIA Technical Blog6565NVIDIA GPUDirect Storage (GDS) accelerates GPU storage access in local and cloud environments, providing a robust, portable, and efficient data path between NVIDIA GPUs and storage targets, increasing bandwidth, minimizing latency, and reducing CPU and memory utilization.
Hugging Face Daily Papers6060Youtu-Parsing-Omni encodes documents, images, audio, and video into spatiotemporal structured records, achieving optimal parsing performance through self-evolution strategies and distillation. Single-modal encoders share weights and integrate cross-modal information, a single architecture unifies outputs for ten parsing tasks. The 5B parameter version achieved top scores of 96.96 and 75.08 on OmniDocBench v1.6 and OmniParsingBench, with code and weights open-sourced.
Hugging Face Daily Papers3131SpecSlot improves pixel space generation Transformer by using local spectral slots and limiting global attention without increasing computational cost, enhancing image details. SpecSlot expands each spatial anchor to multiple slots, assigning each slot a different DCT coefficient band, which separates anchor density and spectral refinement; only the low-frequency slots of each anchor participate in global attention, while all slots perform local interactions.
Microsoft Research4040Microsoft Research has open-sourced an industrial evaluation data generation platform that simulates real business processes by creating synthetic enterprise environments based on O*NET occupational data, online job postings, and SEC filings to test Copilot AI, ensuring the evaluation does not involve customer data and complies with security and privacy standards. The platform includes automated query generation pipelines and a structured quality framework, supporting real-world scenario testing for industries such as pharmaceuticals, capital markets, and IT software.
Same story — showing the featured version《Mistral releases open-source model Mistral Large 4 (Le Chonk) with trillion parameters》
Ant Group inclusionAI · Hugging Face new models3838inclusionAI releases ZwZ2-9B GGUF format weights, belonging to the Zooming without Zooming v2 project, including language models (FP16, Q8_0, Q4_K_M) and visual encoders and projectors (mmproj, FP16, Q8_0, Q4_K_M), requiring pairing with llama.cpp Qwen3.5 multimodal support for use, supporting CLI inference and Web Chat deployment.
Ant Group inclusionAI · Hugging Face new models2222inclusionAI/ZwZ2-27B-GGUF model weights released, including language models and visual components with FP16, Q8_0, Q4_K_M precision, supporting llama.cpp multimodal inference.
MarkTechPost5959Samaya AI releases the Sterling model, fine-tuned from the 30B mixed expert model Nemotron 3.5 Lightning, with 3B active parameters. In the FrontierFinance benchmark test, Sterling scored 58.9%, higher than Claude Fable 5.1's 55.9%, and the cost per question is only $0.61 (Fable 5.1 is $6.89).
NVIDIA and MIT team open source video enhancement solution Open-H3-Regenerate-2K, based on MiniMax H3 model to regenerate low-resolution video to 2K (2560×1440) or 4K (3840×2160) video.
Hugging Face Daily Papers4242WorkFactory builds coherent professional tasks through information planning and repair, producing the WorkPilot model, enhancing expert-level agent performance in professional work benchmarks. The framework starts from professional intent, constructs information support graphs to guide task instantiation, and identifies support failures and guides repairs through execution detection while preserving expected work.
Hugging Face Daily Papers6565LOWN adapts to skip image processing during diffusion visual language model decoding, reducing computational load while maintaining accuracy when predictions are consistent. Since visual markers account for 89-97% of input in the studied benchmarks, most decoding computations are used for repetitive processing of the same images.
Developer Divyansh Agrawal released the experimental model Qwenputer on October 22, built based on Alibaba's Qwen3.5-4B, which implements 32-bit RISC-V CPU and memory functions inside the neural network, allowing program execution and precise calculations without calling external tools like Python.
Why it matters: Developers compiled the RISC-V instruction set into the Qwen3.5-4B weights, demonstrating with specific benchmark numbers the improvement in computational precision within the model and the trade-offs in volume and speed.
MarkTechPost5858Edge0 releases the open-source text-to-speech model Audio8-TTS-Realistic, with 0.6B parameters supporting zero-shot voice cloning for 23 languages, achieving an average error rate of 4.86% in the CV3 benchmark test, surpassing Qwen3-TTS-1.7B and VoxCPM2, but lower than the 4B-level Higgs TTS 3. The model employs a dual autoregressive architecture, including a slow branch with 24 layers and width 896, and a fast parallel prediction branch with 4 layers, trained using a learned realism critic.
