THE NEXT THING, IN PERSPECTIVEIndependent AI news

The latest AI news, in focus.

Official announcements and industry reporting. Follow the timeline of AI.

Today's news

News feed
AI NEWS × FOXAIFind your next step

From the latest news to tools that work for you.

Explore AI tools

Discover AI tools

Swipe to explore →
KiroCoding & development

Kiro is an AI engineering tool for developers that organizes requirements into specifications, architectures, and tasks, generates code via agents, and supports the entire workflow—from planning to code review—through property-based testing, execution memory, and integration with development tools.

AI programming tool分类推荐榜首Free entry availableView details
EvoMapAgents & automation

EvoMap is an infrastructure designed for the reuse of AI agent experiences. It organizes reusable strategies and repair assets with validation records via the GEP protocol, offering market- and quality-based ranking to help different agents discover, evaluate, and inherit existing capabilities.

分类推荐榜首Pricing unverifiedView details
AI ShortWriting & content

AIS ALERT MANAGEMENT AND SHARED PLATFORM, QUICK COMMAND FOR MULTIPLE SCENARIOS

AI HINT COMMAND分类推荐榜首Pricing unverifiedView details
Open DesignDesign & 3D

OpenDesign is an open-source design workspace that can connect to local programming agents, use built-in model keys, or leverage cloud-based models, enabling the creation and editing of prototypes, web pages, presentations, and graphics within a design system; local software is free, while model services incur fees.

AI design tools分类推荐榜首Free entry availableView details
JAXModels & infrastructure

JAX is a Python numerical computing library designed for accelerators, specifically tailored for high-performance array computations and large-scale machine learning. It efficiently executes the same code across multiple hardware backends, including CPUs, GPUs, and TPUs.

AI DEVELOPMENT PLATFORM分类推荐榜首Pricing unverifiedView details
MemoAIMusic & audio

MemoAI is an AI-powered audio-to-text tool that supports localized processing, eliminating the need to upload data and ensuring privacy and security. It can efficiently transcribe YouTube videos, podcasts, and local audio/video files (such as MP4, MP3, AAC, M4A, etc.) into text, while offering core features like multilingual translation, speaker diarization, real-time subtitles, and floating notes—making it ideal for meeting minutes, interview transcription, content learning, and more.

AI AUDIO TOOL分类推荐榜首Pricing unverifiedView details
ElicitKnowledge & research

Elicit provides semantic search, research report generation, literature screening, and data extraction for scientific research. It enables traceable answers within source materials and organizes evidence through a literature library, research agents, and ongoing tracking workflows, making it ideal for advancing from initial problem exploration to systematic reviews.

paperLiteratureResearch待评分 · 分类列表推荐Free entry availableView details
Udacity AI学院Education & languages

Udacity AI Academy is the gateway to AI courses and learning paths, covering machine learning, deep learning, computer vision, natural language processing, and AI product management, with courses and project-based training available at both beginner and advanced levels.

AI learning website待评分 · 分类列表推荐Pricing unverifiedView details
Meta AI助手Assistants & search

Meta's free AI chat assistant

待评分 · 分类列表推荐Pricing unverifiedView details
闪剪Video & animation

FlashCut is an AI-powered creation platform for voice-over and short-video marketing. It can generate finished videos with voiceovers, subtitles, and visuals from text, offering digital human customization, photo-to-digital-human conversion, live-stream clipping, and intelligent editing. It also provides APIs and branded mini-program solutions for businesses.

待评分 · 分类列表推荐Pricing unverifiedView details

AI, as it happens

2401 published stories
Compute & infrastructure · 327 storiesClear filters ×
2026-11-20
Follow the timeline
18:46

b12561

The update introduces a chunked GATED_DELTA_NET implementation for Vulkan, including shader additions, optimizations for RDNA4 and Ampere architectures, fixes for shared memory bank conflicts and NVIDIA shader compiler issues, and support for various subgroup sizes and backend configurations. It also enables the feature by default and adds diagnostic markers for device information.

Inference systems & deploymentRead original
14:38

b12558

A fix has been implemented to disable subgroup arithmetic in the Vulkan driver for Qualcomm's Adreno GPUs, addressing an issue where subgroupAdd() after divergent code returns incorrect sums. This correction resolves problems with the decode-step norm, preventing models from generating stray tokens after long prompts. Testing on Adreno 850 and X2-90 showed improved results for ADD_RMS_NORM and RMS_NORM_MUL_ADD operations, while performance on llama 3B Q4_0 remained unchanged.

Inference systems & deploymentRead original
13:00

Berth: SLO-Preserving GPU Sharing Without New Preemption

arXiv:2611.25091v1 Announce Type: new Abstract: Inference services keep GPUs reserved for their worst-case load, and most of that capacity sits idle. Filling it with best-effort work such as training is the obvious remedy, and latency is why it rarely happens: on a shared GPU, a request that must finish within milliseconds finds the best-effort job's kernels running or queued ahead of it. Spatial sharing cannot evict them, temporal sharing gives the request only its turn, and hardware partitioning protects it by idling what it does not use. Systems that keep both the SLO and the utilization have reached inside the stack (rewriting kernels, patching the driver, instrumenting the framework), which the operator of a closed GPU stack cannot do. This paper argues that reaching inside is unnecessary. The time-slicer built into commodity GPUs already preempts a running kernel within a fraction of a millisecond and switches the moment a context runs out of work; what it lacks is priority, and priority can be created from outside by controlling what the best-effort context has to run. Berth is a user-space system built on this observation: a preloaded shim feeds the best-effort job to the GPU only when no request needs it (feed control), a declared deadline reference extends the rule to continuous-batching LLM engines, and a shortened best-effort time slice, set without privileges, bounds what a long kernel can cost. Berth changes no application, framework, kernel or driver. Across 31 scenario points with MLPerf SLOs and 13 baselines, Berth meets SLOs within a few percentage points of running alone, above every baseline, with 2.7-3.5x the best-effort throughput of the strongest ones; beside LLM serving, it meets per-token targets for nearly every request while training keeps running. Sharing a closed GPU under SLOs needs no new preemption mechanism, only control over what the existing one sees.

Chips, GPUs & hardwareRead original
07:12

b12556

A vulnerability in llama.cpp's HRM layer count validation allows integer overflow due to improper handling of uint32_t multiplication, leading to potential buffer overflows during tensor loading. The issue arises when the product of n_hrm_layers_per_stack, n_hrm_h_cycles, and (n_hrm_l_cycles + 1) is computed as a uint32_t and compared to block_count, enabling maliciously crafted inputs to pass validation despite exceeding the actual block count.

Inference systems & deploymentRead original
05:51

b12552

The update includes fixes for OpenVINO CI, optimizations for model recompiles in the prefill stage, improvements to output handling during model reuse, and support for f16 SCALE and f16/bf16 TRI operations. It also addresses issues with static-shape views in MoE layers, enhances SSM_SCAN build checks, and adds Falcon-H1 to the list of validated OpenVINO models with noted limitations.

Inference systems & deploymentRead original
05:07

Enterprise AI is pushing cloud infrastructure beyond the GPU

Enterprise adoption of artificial intelligence is creating new requirements for enterprise AI infrastructure. While much of the early AI investment centered on access to powerful GPUs, organizations moving AI applications and agents into production are discovering that accelerators are only one part of the equation. Agentic AI systems need environments where they can execute code, […] The post Enterprise AI is pushing cloud infrastructure beyond the GPU appeared first on SiliconANGLE.

Data centres & cloud computeRead original
04:00

NVIDIA and Microsoft Expand Accelerated Computing From Azure to Windows PCs

NVIDIA and Microsoft are extending their full-stack collaboration to help organizations build, customize and deploy intelligent applications — from next-generation infrastructure on Azure to industrial systems and Windows PCs. At Microsoft Ignite, the companies are expanding access to Azure virtual machines powered by NVIDIA Vera Rubin NVL72 systems, as well as adding customer availability to […]

Chips, GPUs & hardwareRead original
02:39

Hitachi Vantara bets smarter storage can close AI’s pilot-to-production gap

Enterprises pushing AI from pilot to production are learning that data, not GPUs, is the hardest part, and open table formats such as Apache Iceberg are part of the answer. Demand for AI infrastructure shows few signs of slowing, but much of the information companies need is still scattered, ungoverned and out of reach. That […] The post Hitachi Vantara bets smarter storage can close AI’s pilot-to-production gap appeared first on SiliconANGLE.

Chips, GPUs & hardwareRead original
01:46

b12549

The llama.cpp v1.25.49 release includes fixes for Inkling 2 attention mask handling, making the relative-position input a hybrid iswa input, enabling optional expert weight scale and final logit softcap, and improving KV rows mask generation for hybrid iswa memory. The update also standardizes code formatting in inkling2.cpp and ensures compatibility across multiple platforms including macOS, Linux, Windows, Android, and openEuler.

Inference systems & deploymentRead original
00:33

b12543

The update introduces a new tensor API conv_2d kernel for Metal, enabling split-K optimizations across all chips, including M5. It improves efficiency by utilizing tensor cores, reducing idle GPU time during long reductions. The kernel now splits reductions into partial sums stored in buffers, with a follow-up kernel summing them deterministically. Address math is optimized using 32-bit byte offsets, reducing computational overhead. A fallback to the generic kernel is implemented for images exceeding 4 GiB. Performance benchmarks show significant speed improvements on M5 Max devices, with reduced execution times across various convolution configurations.

Inference systems & deploymentRead original
2026-11-19
Follow the timeline
19:40

b12535

The Vulkan backend in llama.cpp has been updated to implement stream-k MUL_MAT, which splits data chunks of 256 K elements evenly across all SMs. Workgroups that finish a tile write to the destination, while others write to scratch, with a resolve shader similar to split-k. The feature is implemented for scalar+cm1+cm2 but currently enabled only for cm2. A simple heuristic is used for enabling on coopmat2 due to large tile size, though tuning for other tile sizes is not planned for NVIDIA devices.

Inference systems & deploymentRead original
13:00

Accelerating Distributed Digital Ising Machines

arXiv:2611.23883v1 Announce Type: cross Abstract: Ising machines, dedicated accelerators built around the dynamics of Ising-like spin models, have emerged as a leading platform for combinatorial optimization and probabilistic sampling. The problems that motivate them require millions of spins, far beyond the on-chip capacity of any single device, so a practical Ising machine must distribute its computation across many devices. Prior work reports throughput at a single operating point, leaving the binding resource unidentified and platform crossovers measurable only by experiment; no calibrated cross-platform cost model of distributed Ising machines has been reported, nor have systematic techniques to accelerate their Markov chain Monte Carlo (MCMC) workloads. This work presents a calibrated cost model that decomposes digital Ising sweep time into compute, memory, synchronization, and inter-device communication terms, validated with 4.8% mean error across three GPU generations, FPGA and ASIC up to 1 million spins. Guided by the model, we address each term in turn and propose: (i) a data placement strategy, worth up to 4.2$\times$ speedup; (ii) multi-rate and bounded-staleness schedules that relax the barrier that Gibbs sampling imposes, cutting time-to-solution (TTS) by up to 3.16$\times$ on three benchmarks; and (iii) a phase-offset streaming dataflow that hides boundary exchange behind interior computation whenever a link satisfies a latency-transparency criterion, yielding up to 1.25$\times$/1.74$\times$/2.01$\times$ on an H200 pair, a ZCU106 pair, and a four-board XEM8320 $2{\times}2$ mesh. The streaming dataflow is also implemented on a fabricated four-chip 28 nm ASIC sustaining 28.1 Gflips/s at 4.81 pJ/flip.

Chips, GPUs & hardwareRead original
13:00

BiMa-ECGSR: reconstructing the 12-lead ECG from reduced lead sets with bidirectional state-space models

arXiv:2611.23345v1 Announce Type: new Abstract: Wearables and monitors record few electrocardiogram (ECG) leads; most diagnostic tools assume 12 leads. BiMa-ECGSR reconstructs the 12-lead ECG from reduced lead sets with a bidirectional Mamba state-space encoder--decoder, as regression (BiMa-SR) or diffusion (DiBiMa-SR). Models were trained on PTB-XL for six input sets (I; II; I+II; I+II+V1; I+II+V2; I+II+V1+V4), see only the measured leads at inference, and were tested on the PTB-XL test fold (2,198 records) and two external cohorts (Georgia, 10,275 records; Chapman-Shaoxing/Ningbo, 43,559). From lead II alone, the pooled median Pearson correlation (PCC) over the 11 predicted leads was 0.914 on PTB-XL, against 0.784 for a least-squares linear filter (lead I: 0.906 against 0.789). As limb leads are exactly derivable from I and II, multi-lead inputs were compared on predicted chest leads: 0.939 against 0.849 for I+II and 0.979 against 0.962 for I+II+V1+V4. BiMa-SR exceeded the linear comparator in all 18 configuration--cohort pairs (by 0.012--0.136), a margin that shrank as chest leads were added. From lead II, regression was more accurate than diffusion (DiBiMa-SR: 0.903 one-step, 0.862 with six DDIM steps). A 12-lead classifier trained only on real ECGs reached a superclass macro AUROC of 0.912 on real PTB-XL ECGs, 0.843 on reconstructions from lead II and 0.908 from I+II+V1+V4; externally, reconstructions from three or four leads scored within 0.004 of real ECGs. BiMa-SR (2.68\,M parameters) takes 45.8 ms per 10-s record on a GPU. Reconstruction makes reduced-lead recordings usable by 12-lead tools; it cannot add information absent from the measured leads.

Chips, GPUs & hardwareRead original
13:00

DexAttention: Distributed Sparse Attention through Rapid Online Planning and Persistent GPU Execution

arXiv:2611.23512v1 Announce Type: new Abstract: Sparse attention reduces computation, but distributed execution may still transfer unnecessary data and incur substantial per-operation overhead. Mask-aware assignment and scheduling can reduce these costs, but replanning for dynamic masks must be cheaper than the execution time it saves. We present DexAttention, a distributed sparse attention system that co-designs fast online planning with persistent execution. Its assigner uses GPU-accelerated, topology-aware clustering to balance computational load and reduce communication. A lightweight greedy scheduler orders coarse communication and computation operations, while a persistent kernel on each GPU enforces fine-grained dependencies and dispatches ready tasks. This division keeps detailed synchronization out of planning while enabling overlap without repeated attention launches. The framework supports static and dynamic per-head masks across multiple attention compute kernels. On a four-node, 16-GPU cluster, DexAttention achieves a $1.39\times$ geometric-mean end-to-end training speedup over the fastest applicable baseline across 11 configurations. For dynamic video attention, it achieves $2.11$--$2.37\times$ speedups over Ring Attention including planning.

Chips, GPUs & hardwareRead original
06:56

b12527

The update rejects malformed GGUF files that could cause out-of-bounds memory access by validating the mudd_tap_idx array bounds during hparams loading, preventing potential exploits from crafted models with excessive n_layer * n_mudd_taps values exceeding 512.

Inference systems & deploymentRead original
01:06

Agentic AI puts the CPU back at center stage as Dell and AMD chase every watt

Agentic AI is rewriting the business development playbook for chipmakers and server builders. Graphics processing units still grab the headlines, but as supercomputing and AI converge, the next wave of demand is landing on central processing units that must fit inside the power budgets data centers already have. The shift has been building for years. […] The post Agentic AI puts the CPU back at center stage as Dell and AMD chase every watt appeared first on SiliconANGLE.

Data centres & cloud computeRead original
2026-11-18
Follow the timeline
22:57

AI’s hidden energy tax: More than 80% of scale-up traffic is just memory moving around

Memory bottlenecks are emerging as one of the costliest problems in AI infrastructure as inference overtakes training. GPUs sit idle waiting on data, while high-bandwidth memory, or HBM, stays bolted to processors that cannot share it. Photonics is shaping up as one way out. Marvell Technology Inc. has been pairing optics with photonic memory to […] The post AI’s hidden energy tax: More than 80% of scale-up traffic is just memory moving around appeared first on SiliconANGLE.

Chips, GPUs & hardwareRead original
13:00

AInfer-PD: Communication-Safe In-Place Prefill-Decode Multiplexing for Distributed MoE Rollouts

arXiv:2609.00993v3 Announce Type: replace Abstract: In agentic reinforcement learning (RL), asynchronous multi-turn trajectories repeatedly introduce continuation prefills (P) while other trajectories remain in decode (D). Efficient rollout completion therefore requires D to advance during long P executions. In-place multiplexing shares devices, model weights, and KV state, but phase-specific communication in distributed MoE introduces two constraints: cross-rank collective ordering and ownership of mutable communication state. Whole-prefill coordination can ensure safe ordering but delays D. DeepEP's normal and low-latency paths suit large P and small D batches, respectively, but safely overlapping them requires separating their mutable communication state. We present AInfer-PD, which extends in-place P/D multiplexing to distributed MoE rollouts. AInfer-PD establishes rank-consistent collective ordering at safe communication boundaries, allowing successive D iterations to advance without waiting for an entire prefill execution to be enqueued or completed. It gives P and D separate communication buffers and protocol state within one DeepEP runtime, allowing P to use the normal path while D retains the low-latency path. Together, these mechanisms preserve useful P/D overlap while sharing model weights and KV storage on the same devices. Across repeated single-node prefill-intensive workloads, AInfer-PD reduces fixed-workload rollout completion time by 7.1-22.5% relative to the same AInfer engine with P/D multiplexing disabled and by 24.8-32.9% relative to SGLang. On two nodes, the reductions are 18.0-35.3% and 18.3-31.8%, respectively. In a same-engine ablation, fine-grained boundaries reduce completion time by a further 8.6-19.8% over whole-prefill asynchronous enqueue.

Inference systems & deploymentRead original
Every update links back to its source.
STAY CURIOUS. STAY INFORMED.

Stay informed about the changes that matter.

Give your curiosity direction. Make every day of reading count.

Subscribe via RSS