Search “Reasoning”

Tencent Hunyuan Releases Hy-Nomas Technical Report

On October 30, 2026, Tencent Hunyuan released the technical report for Hy-Nomas, an interactive video world model, and open-sourced Hy-Nomas-Bench for reproducible research. The official team also reproduced Genie 4's examples using Hy-Nomas and released comparison videos, and provided an online demo with no waitlist. Earlier introductions stated that Hy-Nomas supports free interaction with text prompts and camera trajectories, long-horizon generation with world memory, lightweight LoRA adaptation for fields such as robotics, and optimization for inference on consumer-grade GPUs. There are currently no further developments.

Ceva announces it will hold an edge technology seminar in Shanghai on November 5

Ceva announced that the 2026 Ceva Edge Technology Seminar will be held on November 5 at the Hilton Shanghai Zhangjiang, aimed at semiconductor engineers, SoC architects, and embedded developers. According to Jiwei.com, the event is divided into two tracks, AI and sensing, and connectivity technology, with topics covering memory and bandwidth bottlenecks in LLM inference, NPU architectures for generative and multimodal AI, AI applications below 1 TOPS, as well as Bluetooth 7, Wi-Fi 7/8, UWB, and 5G NTN custom chips. There is currently no further reporting on agenda updates or registration status.

Micro-Nano Core completes over 1 billion yuan Series C2 financing

Hangzhou-based compute-in-memory AI chip company Micro-Nano Core has completed a Series C2 financing round of over 1 billion yuan, led by Jiantou Investment, with the funds to be used for product development and iteration as well as supply chain mass-production stocking. Combined with the Series C1 round completed in September, the company has raised over 2 billion yuan within two months and a cumulative 3 billion yuan over half a year. Its 3D-CIM architecture targets large-model inference; all three of its chips for edge and cloud sides have been taped out, with customer sampling and the start of volume supply expected in early 2027.

Moore Threads Releases 2026 Third Quarter Report

Moore Threads released its 2026 third-quarter report on October 30, 2026. According to TMTPost·AGI, its third-quarter revenue was 1.552 billion yuan, up 1773.92% year on year; non-GAAP net profit attributable to the parent company was 20.0724 million yuan, turning positive on a quarterly basis for the first time, with a non-GAAP net margin of about 1.29%. Cumulative revenue for the first three quarters was 3.289 billion yuan, while non-GAAP net profit attributable to the parent company remained a loss of about 131 million yuan, and net cash flow from operating activities was -3.309 billion yuan. Caixin's morning briefing that day also mentioned that Moore Threads' third-quarter revenue was 1.552 billion yuan, up 1773.92% year on year, and said net profit attributable to the parent company was 49.2604 million yuan, turning from loss to profit year on year. The report also compared the performance of peers such as Cambricon, Hygon Information, and MetaX, noting that the share of inference demand is rising and that HBM and foundry capacity remain the main constraints on the growth of domestic GPUs. There is currently no further follow-up progress.

Eclypsium demonstrates that compromised hardware can tamper with AI models at runtime

Eclypsium demonstrated how compromised computing infrastructure can alter AI model behavior during inference. The experiment was completed in an NVIDIA Grace Hopper environment with NVIDIA's support; attackers used a compromised DPU and lax memory isolation to modify runtime model weights during inference, changing the model's answers while the model files on disk and hash verification still appeared normal. The experiment relied on conditions such as IOMMU passthrough and lax PCIe isolation, which the researchers explicitly noted are not Linux default settings. The report brings the DPU, GPU, management controller and their firmware into the AI security boundary, and notes that model signing, secure boot, runtime isolation and continuous monitoring each answer different questions. There are currently no further developments.

Kakao Open-Sources Two Korean-Optimized Embedding Models

On October 29, Kakao open-sourced two Korean-optimized embedding models on Hugging Face: Kanana-2-3b-embedding (3 billion parameters) and Kanana-2-1.3b-embedding (1.3 billion parameters), targeting personalized AI agents and RAG scenarios. Kakao said that in internal evaluations, the two models outperformed llama-nemotron-embed-1b-v2 and Qwen3-Embedding-4B of comparable scale in Korean retrieval performance, with comparable English performance and higher inference efficiency, while memory usage and inference speed improved by more than tenfold compared with the previous-generation Kanana embedding models. The models adopt MRL, allowing developers to adjust the embedding vector size, and are released under the Kanana Open License for commercial use. There are currently no further developments.

Microsoft opens Copilot agent harness public preview through Foundry

On October 30, 2026, Microsoft opened the public beta of the GitHub Copilot agent harness through the Foundry Agent Service in Microsoft Foundry. The harness provides built-in sandboxing, reusable skills, MCP integration, and sub-agent support; the same harness already supports Copilot Code, Copilot Studio, GitHub Copilot, CLI, and VS Code. When creating a prompt agent, this harness must be selected and cannot be changed afterward; CPU, memory, and idle duration are controlled by Microsoft, and bring-your-own-model is not supported. The service is free until December 1, 2026, after which it is billed at $0.0994 per vCPU-hour and $0.0118 per GiB-hour, with model inference and some tools billed separately.

The Institute for Advanced Study in Mathematics of Zhejiang University and Zhijiang Laboratory launch the Future Mathematics Program

On October 30, 2026, ScienceNet published an interview report mentioning that the Institute for Advanced Study in Mathematics of Zhejiang University and Zhijiang Laboratory jointly launched the Future Mathematics Program. The report stated that the program focuses on four areas: mathematical foundation models, formal verification, AI mathematical mechanisms, and interdisciplinary empowerment. Li Jianshu, an academician of the Chinese Academy of Sciences, said in the report that AI may change the paradigm of mathematical research, and pointed out that mathematics has done far from enough for AI; Wang Jian, an academician of the Chinese Academy of Engineering, believes that AI's reasoning ability currently comes mainly from code, and that learning deeper reasoning from mathematical data will bring greater gains. The report also mentioned that Google DeepMind's AI agent solved 9 previously unsolved Erdős problems and tackled 44 open mathematical conjectures in May this year, and that OpenAI recently published 722 mathematical manuscripts generated by its internal AI models on GitHub, while 25 Fields Medalists including Terence Tao and Deng Yu jointly issued an open letter warning of a serious misalignment between AI and mathematics. Currently, no further progress on the program has been disclosed.

Anthropic interpretability team develops J++ lens

In the Fellows Program project, Kola Ayonrinde of the Anthropic interpretability team led the proposal of J++ lens, an improved version of J-lens that can replace the original method as a plug-and-play drop-in. The work clusters activations layer by layer and fits a Jacobian for each, then combines them with learned weights, while also introducing R-lens's layer-wise relevance propagation and post-hoc filtering of spurious readout tokens. The team says that in evaluations it reads out the intermediate latent variables used in the model's reasoning more reliably than J-lens and R-lens, and has released the code, the lens, and an interactive Neuronpedia demo.

Amazon Science and Booking.com use KD plus Medusa to accelerate travel LLMs

Amazon Science and Booking.com combined knowledge distillation with Medusa technology to optimize Booking.com's travel large language model. The approach compresses a 70B-parameter Llama-3.1 teacher model into a 1.1B TinyLlama student model, maintaining the teacher model's quality while increasing inference speed by 10 to 20 times and reducing costs to less than one-third. The solution has been applied to three features—AI Trip Planner, Smart Filters, and accommodation description generation—and won the Best Paper Award in the Industry Track at ACL.

Elorian and Mercor Release DecoyCount Benchmark

On October 30, 2026, Elorian and Mercor jointly released the DecoyCount benchmark to test whether models are truly counting objects or relying on visual shortcuts. According to Latent Space, frontier models all scored below 50% accuracy on this benchmark, with the highest being GPT-6 Astra at 43.17%. Elorian said its visual reasoning model Glimpse scored 50.48% in a single run on 1,660 questions, making it the only model to exceed 50%, but that result came from internal testing and did not use an agent or tool framework, whereas the six models in the paper each took the average of three runs. Elorian CEO Andrew Dai said the benchmark examines both vision and reasoning, and can distinguish whether a model failed to see the objects or miscounted them.

inclusionAI open-sources the ZwZ model family and the ZoomBench benchmark

On October 30, 2026, Ant Group's inclusionAI open-sourced the ZwZ model family and the ZoomBench fine-grained perception benchmark as companion results to an ICML 2026 paper. According to the introduction of inclusionAI's new GitHub repository, the ZwZ model family includes three sizes—4B, 7B, and 8B—and achieves SOTA on multimodal perception benchmarks among open-source models, requiring only a single forward inference pass and no tool calls at inference time. Its method is Region-to-Image Distillation (R2I), which distills high-quality VQA supervision generated from enlarged local regions back into the full image, with 37K training samples. The simultaneously released ZoomBench contains 845 VQA items covering 6 fine-grained perception dimensions, and Qwen2.5-VL-7B achieves only 42.5% average accuracy. There are currently no further developments.

SanDisk says customers underestimate NAND capacity needed for AI infrastructure

Speaking on its fiscal 2027 first-quarter earnings call, SanDisk said customers have underestimated the NAND capacity needed to run AI infrastructure, and that AI inference is reshaping the supply and demand landscape for NAND flash memory. SanDisk noted that its order pipeline under new business models extends to 2030 and beyond, with about 50% of bit shipments for fiscal 2027 already locked in under NBM agreements, a figure expected to rise to about two-thirds in fiscal 2028. Earnings for the same period showed SanDisk's quarterly revenue at $11.031 billion, up 378% year over year, with data center business revenue of $4.802 billion, accounting for about 43.5% of total revenue. SanDisk also announced that its first high-bandwidth flash (HBF) memory die has completed tape-out and entered the manufacturing stage, with the first samples planned for delivery in the next calendar year.

Physical Intelligence in talks to raise at least $1 billion in new funding

According to Bloomberg, the U.S. robotics AI company Physical Intelligence is in talks for a new funding round of at least $1 billion, with existing investors Avenir and Lux Capital considering leading the round, at a post-money valuation expected to be about $20 billion. The company was founded in 2024 and was valued at $5.6 billion in its November funding round last year; it currently has no commercial revenue and has not mass-produced any robotic products. Its technical approach is to build on VLMs by introducing action-generation capabilities to train VLA models, and it has iterated through π0, π0.5, and π0.7. This month it launched a beta version of a robot foundation model API, offering external developers model fine-tuning and inference services.

LimX Dynamics open-sources the FluxVLA Engine engineering foundation

On October 28, 2026, LimX Dynamics open-sourced the FluxVLA Engine embodied engineering foundation. On the same day, QbitAI reported that LimX Dynamics had updated the second version of the technical report for this open-source engineering foundation, adding online reinforcement learning post-training experiments compared with the first version in September, and expanding the coverage of simulation evaluation and real-robot deployment. The report shows that after online reinforcement learning, SmolVLA's success rates on the ten tasks of LIBERO-10 and RoboTwin Clean increased by 10.0 and 15.4 percentage points, respectively; the inference frequency of 8 model-and-device combinations increased to 2.31–9.64 times. On the real-robot side, the deployment pipeline from model checkpoint to closed-loop execution was validated on ALOHA, the dual-arm Franka, Rebot, and the self-developed humanoid robot LimX Oli. The engineering foundation is currently open-sourced, accompanied by the second version of the technical report.

AWS introduces enterprise-grade Agentic AI platform blueprint based on AgentCore

On October 29, 2026, AWS released an enterprise-grade Agentic AI platform blueprint based on Amazon Bedrock and AgentCore. The blueprint adopts a federated operating model: a central Platform team provides governed shared services, while each Workstream team retains ownership of its own agents, tools, and data. The platform enforces model admission, regional restrictions, rate and token quotas, and budget controls through an independent inference gateway and MCP tool gateway, and uses an evaluation pipeline to perform gate checks before agents go live. There are currently no subsequent developments.

Ant Ling Open-Sources Ling-3.1-flash-1019

On October 27, 2026, Ant Ling open-sourced Ling-3.1-flash-1019. According to Artificial Analysis, the model scored 44 on the Artificial Analysis Intelligence Index v4.3, higher than its predecessor's 41; it is a reasoning model with 560B total parameters and 25B activated parameters, supports a 1M token context window, is released under the MIT license, is available as open weights on Hugging Face, and can be accessed through InclusionAI's first-party API. QbitAI reported that this version was further trained and optimized on the basis of Ling-3.1-flash, and that the base model Ling-3.1-flash-base was also open-sourced. On October 30, Ant Ling announced that the model is being made available to more developers through Novita AI, stating that its reasoning capabilities and real-world agent performance are stronger.

OpenAI launches GPT-6.1 Terra on ChatGPT Work and Codex

On October 27, 2026, OpenAI released and launched GPT-6.1 Terra, touted for its speed and efficiency, visual reasoning, and precise computer operation, with clearer and more natural writing as well, rolling it out that day to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex. On the same day, Arena announced that the model had entered Agent Arena and Code Arena; OpenRouter launched the model, saying it matches GPT-6 Astra on Agents' Last Exam at one-tenth the price per token; OpenClaw also announced that the model is now selectable. OpenAI co-founder Greg Brockman said Terra's computer use has crossed the threshold of a qualitative change and that using it is faster than operating a computer on his own. fal, meanwhile, used the fal Codex plugin to test turning Blender 3D content into Seedance 2.5 cinematic footage. On October 28, Arena announced that Terra ranked 14th in Code Arena: WebDev with a score of 1671, at a median cost of $0.40 per task, 28% lower than GPT-5.6 Terra. OpenAI's system card confirms that Terra is overall safer than older models, but its flag rates on items such as concealing uncertainty, circumventing restrictions, data exfiltration, and destructive operations

NVIDIA and Meta Announce Expanded Collaboration on Muse Models

On October 27, 2026, NVIDIA and Meta announced an expansion of their collaboration on Muse and the Muse series of models to improve inference and training performance, with Meta's Muse models and agents running on NVIDIA infrastructure. According to an announcement published by the NVIDIA Newsroom that day, Meta plans to expand deployment of NVIDIA Vera Rubin and Vera CPU and to adopt Spectrum-X Ethernet and BlueField-4; the two sides will also evaluate the NVIDIA OpenShell reference architecture and NVIDIA Sentry for secure agent deployment and behavior monitoring. The same announcement stated that NVIDIA will invest $3 billion in Meta's equity offering. 36Kr AI reported on October 28 that this investment is part of Meta's total $95 billion fundraising plan, with AMD and Broadcom also expressing interest in investing; Meta expects full-year capital expenditure to reach $135 billion to $145 billion. The collaboration and investment plans have now been announced, while subsequent deployment progress remains to be disclosed.

Eclypsium demonstrates that compromised hardware can tamper with AI models at runtime

Eclypsium demonstrated how compromised computing infrastructure can alter AI model behavior during inference. The experiment was completed in an NVIDIA Grace Hopper environment with NVIDIA's support; attackers used a compromised DPU and lax memory isolation to modify runtime model weights during inference, changing the model's answers while the model files on disk and hash verification still appeared normal. The experiment relied on conditions such as IOMMU passthrough and lax PCIe isolation, which the researchers explicitly noted are not Linux default settings. The report brings the DPU, GPU, management controller and their firmware into the AI security boundary, and notes that model signing, secure boot, runtime isolation and continuous monitoring each answer different questions. There are currently no further developments.

View story:Eclypsium demonstrates that compromised hardware can tamper with AI models at runtime