# acceleration

Published articles for acceleration.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Vulkan Rendering Is Now The Default on Linux Wayland

DevFeed: [Vulkan Rendering Is Now The Default on Linux Wayland](<https://devfeed.tech/articles/vulkan-rendering-is-now-the-default-on-linux-wayland-146494.md>)

Original publisher: [Read original article](<https://blog.jetbrains.com/platform/2026/11/vulkan-rendering-is-now-the-default-on-linux-wayland/>)

Author: Ekaterina Valeeva

Published: 2026-11-20T16:05:48Z

Content type: release

Language: en

Sources: [JetBrains](<https://devfeed.tech/sources/the-jetbrains-blog.md>)

Topics: [GPU rendering](<https://devfeed.tech/topics/gpu-rendering.md>), [Wayland](<https://devfeed.tech/topics/wayland.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [IntelliJ IDEA](<https://devfeed.tech/topics/intellij-idea.md>)

Tags: [2026-3](<https://devfeed.tech/tags/2026-3.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [bug-fix](<https://devfeed.tech/tags/bug-fix.md>), [color](<https://devfeed.tech/tags/color.md>), [compatibility](<https://devfeed.tech/tags/compatibility.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [crashes](<https://devfeed.tech/tags/crashes.md>), [desktop](<https://devfeed.tech/tags/desktop.md>), [development](<https://devfeed.tech/tags/development.md>), [display](<https://devfeed.tech/tags/display.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [drivers](<https://devfeed.tech/tags/drivers.md>), [ides](<https://devfeed.tech/tags/ides.md>), [jetbrains](<https://devfeed.tech/tags/jetbrains.md>), [jetbrains-runtime](<https://devfeed.tech/tags/jetbrains-runtime.md>), [linux](<https://devfeed.tech/tags/linux.md>), [rendering](<https://devfeed.tech/tags/rendering.md>), [vulkan](<https://devfeed.tech/tags/vulkan.md>), [wayland](<https://devfeed.tech/tags/wayland.md>), [x11](<https://devfeed.tech/tags/x11.md>), [xwayland](<https://devfeed.tech/tags/xwayland.md>)

### AI overview

JetBrains IDEs enable Vulkan rendering by default in native Linux Wayland sessions starting with 2026.3, shifting more UI drawing from the CPU to the GPU while retaining Wayland behavior; no configuration is required. Performance and power benefits vary by GPU, driver, desktop environment, and workload. Remote Development is not yet supported, with compatibility work planned for a 2026.3 bugfix release. Some configurations may still show missing primitives, blurry text at fractional scaling, presentation problems, or other rendering issues. Users who encounter problems can disable Vulkan with `-Dsun.java2d.vulkan=false` and report them with IDE/build, system and GPU details, logs, reproduction steps, and visual evidence. X11, XWayland, and other operating systems are unchanged.

### Source excerpt

JetBrains IDEs are adopting a new rendering path on Linux. As you may remember, our IDEs started running in native Wayland mode in Linux Wayland sessions with 2026.1. Starting from 2026.3, the Vulkan rendering pipeline is enabled by default in this setup. Nothing changes for X11, XWayland, or other operating systems. Vulkan rendering is not [...]

## LoongForge-Embodied: A Unified Framework for Training Heterogeneous Embodied Models

DevFeed: [LoongForge-Embodied: A Unified Framework for Training Heterogeneous Embodied Models](<https://devfeed.tech/articles/loongforge-embodied-a-unified-framework-for-training-heterogeneous-embodied-models-145515.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25460>)

Author: LoongForge Team

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Machine Perception](<https://devfeed.tech/topics/machine-perception.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [data](<https://devfeed.tech/tags/data.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [ddp](<https://devfeed.tech/tags/ddp.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [fp8](<https://devfeed.tech/tags/fp8.md>), [framework](<https://devfeed.tech/tags/framework.md>), [fsdp](<https://devfeed.tech/tags/fsdp.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [i-o](<https://devfeed.tech/tags/i-o.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [layer](<https://devfeed.tech/tags/layer.md>), [memory](<https://devfeed.tech/tags/memory.md>), [native](<https://devfeed.tech/tags/native.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

LoongForge-Embodied uses a shared abstraction layer to make training infrastructure reusable across heterogeneous VLA and WAM architectures without constraining model-specific execution needs. The native-PyTorch framework combines shared and model- or hardware-specific optimizations for computation, memory, data I/O, execution, and distributed communication, including FP8 communication for DDP and FSDP. It also includes infrastructure for synthesizing robot data from egocentric data and concurrent simulator evaluation. Across representative models, the paper reports 1.47x-4.38x higher end-to-end training throughput than corresponding open-source baselines and 4.1x-23.2x faster evaluation under the same protocols.

### Source excerpt

arXiv:2611.25460v1 Announce Type: new Abstract: Training Vision-Language-Action (VLA) models and World-Action Models (WAMs) is computationally expensive, while their heterogeneous architectures and fragmented training stacks make system optimizations difficult to reuse across models. To address these challenges, we present LoongForge-Embodied, a native-PyTorch framework that unifies and accelerates the training of heterogeneous embodied models. At its core, LoongForge-Embodied introduces a common abstraction layer that decouples model-specific implementations from shared training infrastructure, enabling cross-model reuse without constraining individual architectures or execution requirements. Buildingon this unified infrastructure, the framework combines reusable acceleration mechanisms with model-specific and hardware-aware optimization strategies across computation, execution, memory, data I/O, and distributed communication, including FP8-based communication optimizations for DDP and FSDP. Beyond training acceleration, LoongForge-Embodied integrates a unified data infrastructure with a pipeline for synthesizing robot data from egocentric data, and an efficient evaluation framework that supports concurrency and optimized computation in simulator. Experiments across representative VLA and WAM models demonstrate 1.47x-4.38x end-to-end training throughput improvements over their corresponding open-source baselines, as well as 4.1x-23.2x faster evaluation under the same protocols.

## AudioTurbo: Fast Text-to-Audio Generation with Rectified Diffusion

DevFeed: [AudioTurbo: Fast Text-to-Audio Generation with Rectified Diffusion](<https://devfeed.tech/articles/audioturbo-fast-text-to-audio-generation-with-rectified-diffusion-145302.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2505.22106>)

Author: Junqi Zhao, Jinheng Zhao, Haohe Liu, Yun Chen, Lu Han, Haoyang Zhang, Xubo Liu, Mark Plumbley, Wenwu Wang

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Generative AI](<https://devfeed.tech/topics/generative-ai.md>), [generative-ai-tools](<https://devfeed.tech/topics/generative-ai-tools.md>), [voice ai](<https://devfeed.tech/topics/voice-ai.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [audio](<https://devfeed.tech/tags/audio.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-sd](<https://devfeed.tech/tags/cs-sd.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [eess-as](<https://devfeed.tech/tags/eess-as.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [generation](<https://devfeed.tech/tags/generation.md>), [inference-speed](<https://devfeed.tech/tags/inference-speed.md>), [tta](<https://devfeed.tech/tags/tta.md>)

### AI overview

AudioTurbo accelerates text-to-audio generation by integrating rectified diffusion into pre-trained diffusion models. It learns first-order approximate ODE paths from deterministic noise-sample pairs produced by a pre-trained model, making use of those models rather than relying on a dedicated flow-based generative framework. The reported experiments find that AudioTurbo exceeds a flow-matching-based acceleration model in both generation quality and inference speed; the excerpt provides no quantitative results or further implementation details.

### Source excerpt

arXiv:2505.22106v2 Announce Type: replace-cross Abstract: Diffusion models have significantly improved the quality and diversity of text-to-audio (TTA) generation but are hindered by slow inference speed. Recent flow-based approaches have been explored for accelerating TTA generation. However, these approaches typically rely on dedicated flow-based generative frameworks and do not directly exploit existing pre-trained diffusion-based TTA models. To address this limitation while leveraging the strong generative capabilities of pre-trained diffusion models, we integrate rectified diffusion into these models to improve TTA generation efficiency. Specifically, we propose AudioTurbo, which learns first-order approximate ordinary differential equation (ODE) paths from deterministic noise-sample pairs generated by a pre-trained TTA model. Experiments show that our approach outperforms a flow-matching-based acceleration model in both generation quality and inference speed.

## FlashX: Scaling Robotic Experience Reuse for Embodied Agents with CBA-Based NAND Flash

DevFeed: [FlashX: Scaling Robotic Experience Reuse for Embodied Agents with CBA-Based NAND Flash](<https://devfeed.tech/articles/flashx-scaling-robotic-experience-reuse-for-embodied-agents-with-cba-based-nand-flash-145540.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25580>)

Author: Yuseon Choi, Sanghyeon Park, Eunjong Lee, Hanseo Lee, Sangjin Kim

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Humanoid Robots](<https://devfeed.tech/topics/humanoid-robots.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [diffusion-transformers](<https://devfeed.tech/topics/diffusion-transformers.md>), [Cache](<https://devfeed.tech/topics/cache.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [acquisition](<https://devfeed.tech/tags/acquisition.md>), [actions](<https://devfeed.tech/tags/actions.md>), [agents](<https://devfeed.tech/tags/agents.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [array](<https://devfeed.tech/tags/array.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [banks](<https://devfeed.tech/tags/banks.md>), [based](<https://devfeed.tech/tags/based.md>), [better](<https://devfeed.tech/tags/better.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [experience](<https://devfeed.tech/tags/experience.md>), [flash](<https://devfeed.tech/tags/flash.md>), [nand](<https://devfeed.tech/tags/nand.md>), [nand-flash](<https://devfeed.tech/tags/nand-flash.md>), [reuse](<https://devfeed.tech/tags/reuse.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

FlashX uses CBA-based NAND flash to scale retrieval of successful robotic experiences that warm-start VLA and WAM action generation, aiming to improve adaptation to unfamiliar environments while reducing inference cost. It combines parallel pSLC key search using interleaved Compute-in-Flash engines with dense TLC latent storage; proprioception-aware clustering reduces search and supports incremental additions without global reclustering. Across the evaluated workloads, task success improved by up to 12.24 percentage points in the dense configuration and 7.43 points in the compute-efficient configuration. FlashX reports 2.57x the search throughput of the strongest prior NAND search architecture under pSLC sensing, with clustering reducing search a further 9.7x. Compared with SSD-based retrieval, it reports 1.29-3.95x end-to-end speedup and 1.12-2.34x energy efficiency improvement.

### Source excerpt

arXiv:2611.25580v1 Announce Type: cross Abstract: Embodied agents are emerging as a promising paradigm for complex robotic tasks, leveraging Vision-Language-Action (VLA) and World Action Models (WAMs) to generate actions through Diffusion Transformer (DiT)-based iterative denoising. However, existing models often struggle to generalize to out-of-distribution (OOD) environments. We observe that retrieving successful experiences and using them to warm-start denoising enables agents to better adapt to unfamiliar situations while reducing computational overhead. This creates a new opportunity to improve task accuracy and inference efficiency simultaneously, in contrast to conventional acceleration techniques that trade accuracy for computation. However, realizing this opportunity requires scaling experience banks, introducing substantial storage and search overhead that challenges the capacity, bandwidth, and power constraints of edge systems. We propose FlashX, a CMOS Directly Bonded to Array (CBA)-based NAND flash architecture for scalable experience reuse in embodied agents. FlashX exploits CBA's high internal bandwidth for parallel pSLC key search, employing interleaved Compute-in-Flash (CIF) engines to sustain sensing throughput while retaining dense TLC latent storage. Proprioception-aware clustering further reduces search overhead and enables incremental experience acquisition without global reclustering. Evaluations across VLA and WAM workloads demonstrate up to 12.24\%p and 7.43\%p improvements in task success under dense and compute-efficient configurations, respectively. FlashX achieves 2.57$\times$ higher search throughput than the strongest prior NAND search architecture under pSLC sensing, with a further 9.7$\times$ search reduction through clustering. This delivers 1.29--3.95$\times$ end-to-end speedup and 1.12--2.34$\times$ energy efficiency improvement over SSD-based retrieval.

## Meet the Zephyr Community at Open Source Summit Japan 2026

DevFeed: [Meet the Zephyr Community at Open Source Summit Japan 2026](<https://devfeed.tech/articles/meet-the-zephyr-community-at-open-source-summit-japan-2026-144769.md>)

Original publisher: [Read original article](<https://www.zephyrproject.org/meet-the-zephyr-community-at-open-source-summit-japan-2026/>)

Author: Susan Remmert

Published: 2026-11-19T05:00:16Z

Content type: news

Language: en

Sources: [Zephyr Project](<https://devfeed.tech/sources/zephyr-project.md>)

Topics: [Zephyr RTOS](<https://devfeed.tech/topics/zephyr-rtos.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [coding-community](<https://devfeed.tech/topics/coding-community.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [2027](<https://devfeed.tech/tags/2027.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [ai](<https://devfeed.tech/tags/ai.md>), [amd](<https://devfeed.tech/tags/amd.md>), [blog](<https://devfeed.tech/tags/blog.md>), [community](<https://devfeed.tech/tags/community.md>), [contributors](<https://devfeed.tech/tags/contributors.md>), [developers](<https://devfeed.tech/tags/developers.md>), [event](<https://devfeed.tech/tags/event.md>), [events](<https://devfeed.tech/tags/events.md>), [experts](<https://devfeed.tech/tags/experts.md>), [industry](<https://devfeed.tech/tags/industry.md>), [industry-conference](<https://devfeed.tech/tags/industry-conference.md>), [innovation](<https://devfeed.tech/tags/innovation.md>), [japan](<https://devfeed.tech/tags/japan.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [maintainers](<https://devfeed.tech/tags/maintainers.md>), [meetup](<https://devfeed.tech/tags/meetup.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [open-source-summit](<https://devfeed.tech/tags/open-source-summit.md>), [solutions](<https://devfeed.tech/tags/solutions.md>), [tokyo](<https://devfeed.tech/tags/tokyo.md>), [zephyr](<https://devfeed.tech/tags/zephyr.md>)

### AI overview

Open Source Summit Japan 2026 is scheduled for December 7-9 at Toranomon Hills Forum in Tokyo, with a Zephyr track on December 8-9 in Hall B3. Sessions are described on topics including TSN, PCIe, safety certification and testing, Edge AI, spacecraft OTA, API contracts, AMD Versal memory architecture, GPU acceleration, multimedia, and displays. The Zephyr Project plans to staff Booth 27 with live demonstrations, including Edge AI, Bluetooth Channel Sounding, space applications, and Zephyr on a Zaurus PDA; additional demos are still being planned. The event also marks Zephyr's 10th anniversary. A community meetup is planned, but its date, venue, agenda, and registration details have not yet been announced.

### Source excerpt

The Zephyr community is heading to Tokyo this December for Open Source Summit Japan 2026! Taking place from December 7-9 at Toranomon Hills Forum in Tokyo, Japan, the event brings together open source developers, contributors, maintainers, technologists, community leaders, and industry experts to share knowledge, collaborate on solutions, and advance open source innovation.

## FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment

DevFeed: [FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment](<https://devfeed.tech/articles/flexmoe-learning-one-for-all-multi-budget-expert-pruning-for-flexible-moe-deployment-143284.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.27866>)

Author: Fan Mo, Yuxuan Han, Geng Zhang, Wangbo Zhao, Yang You

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [Post-training optimization](<https://devfeed.tech/topics/post-training-optimization.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [actions](<https://devfeed.tech/tags/actions.md>), [activation](<https://devfeed.tech/tags/activation.md>), [budget](<https://devfeed.tech/tags/budget.md>), [co-design](<https://devfeed.tech/tags/co-design.md>), [compression](<https://devfeed.tech/tags/compression.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [experts](<https://devfeed.tech/tags/experts.md>), [ffn](<https://devfeed.tech/tags/ffn.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [flexible](<https://devfeed.tech/tags/flexible.md>), [inference](<https://devfeed.tech/tags/inference.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [learning](<https://devfeed.tech/tags/learning.md>), [llms](<https://devfeed.tech/tags/llms.md>), [masks](<https://devfeed.tech/tags/masks.md>), [memory](<https://devfeed.tech/tags/memory.md>), [moe](<https://devfeed.tech/tags/moe.md>), [pruning](<https://devfeed.tech/tags/pruning.md>)

### AI overview

FlexMoE aims to produce deployable MoE subnetworks at multiple parameter budgets from one pretrained model, avoiding separate compression runs for each budget. It ranks expert FFN channels to create a shared sliceable weight space, then learns expert-wise discrete retention actions under increasing cost pressure. A single fine-tune at a mid-range budget can recover performance and transfer to unseen budgets, reducing repeated recovery and separate checkpoint storage. The abstract reports gains over tested compression baselines; on Qwen2-57B-A14B, pruning 50% of routed-expert parameters retained about 99.8% of base performance without recovery fine-tuning. It also reports memory reduction and throughput gains, with kernel co-design enabling runtime budget switching without reloading checkpoints.

### Source excerpt

arXiv:2606.27866v2 Announce Type: replace Abstract: Mixture-of-Experts (MoE) language models scale model capacity through sparse expert activation, yet deployment still requires storing and serving all experts. Moreover, deployment budgets can vary substantially across devices, users, and workloads. Existing MoE compression methods typically optimize a separate compressed model for each target budget. We introduce FlexMoE, a post-training one-for-all framework that converts a large pretrained MoE model into a multi-budget family of smaller deployable subnetworks. FlexMoE first ranks expert FFN channels by importance to construct a sliceable weight space shared across budgets, and then learns expert-wise discrete retention actions under gradually increasing cost pressure, allowing a single action-training run to produce reliable pruning masks across budgets. Moreover, only a single mid-budget fine-tune can further recover degraded performance and yields a shared recovered model that transfers to other unseen budgets, avoiding repeated per-budget recovery and separated checkpoint storage. Across tested MoE LLMs with diverse architectures and scales, FlexMoE consistently outperforms strong MoE compression baselines. In particular, on Qwen2-57B-A14B, it retains ~99.8% of base performance while pruning 50% of routed-expert parameters even without recovery fine-tuning. For deployment, pruned subnetworks deliver real memory reduction and end-to-end throughput gains, while our kernel co-design enables runtime budget switching without checkpoint reloading and delivers substantial inference acceleration.

## Efficient simulation of variational quantum circuits: representation, accuracy and CPU/GPU allocation

DevFeed: [Efficient simulation of variational quantum circuits: representation, accuracy and CPU/GPU allocation](<https://devfeed.tech/articles/efficient-simulation-of-variational-quantum-circuits-representation-accuracy-and-cpu-gpu-allocation-142982.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24444>)

Author: Vojt\\v{e}ch Nov\\'ak, Radovan Re\\v{c}ka, Ond\\v{r}ej Knebl, Ivan Zelinka

Published: 2026-11-19T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [Simulation](<https://devfeed.tech/topics/simulation.md>), [Quantum Computing](<https://devfeed.tech/topics/quantum-computing.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [accelerate](<https://devfeed.tech/tags/accelerate.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [based](<https://devfeed.tech/tags/based.md>), [benefits](<https://devfeed.tech/tags/benefits.md>), [can](<https://devfeed.tech/tags/can.md>), [change](<https://devfeed.tech/tags/change.md>), [circuit](<https://devfeed.tech/tags/circuit.md>), [circuit-simulation](<https://devfeed.tech/tags/circuit-simulation.md>), [circuits](<https://devfeed.tech/tags/circuits.md>), [compare](<https://devfeed.tech/tags/compare.md>), [complete](<https://devfeed.tech/tags/complete.md>), [complex](<https://devfeed.tech/tags/complex.md>), [compression](<https://devfeed.tech/tags/compression.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [quant-ph](<https://devfeed.tech/tags/quant-ph.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [quantum-circuits](<https://devfeed.tech/tags/quantum-circuits.md>), [simulation](<https://devfeed.tech/tags/simulation.md>)

### AI overview

The study compares statevector simulation, direct Pauli evolution, and matrix product states for molecular and spin variational Hamiltonian ansätze and hardware-efficient circuits across CPU/GPU systems. In the tested implementations, direct Pauli evolution benefits complex molecular circuits, while GPUs accelerate wider spin and hardware-efficient circuits. Matrix product states can reduce cost when entanglement is limited enough for accurate compression. Corrected CPU FP32 met the tested tolerance but cost more than FP64. Allocation choices depend on workload size, startup costs, worker memory, and profiling: matched NumPy Pauli updates sped up two-layer full LiH by 3.45x, and a GPU was about 30x faster than 18-thread Aer for a 28-qubit, four-layer TFIM, both in FP64. The proposed procedure jointly selects representation, precision, and worker allocation by comparing complete energy-evaluation costs under accuracy, memory, and core constraints.

### Source excerpt

arXiv:2611.24444v1 Announce Type: cross Abstract: Classical circuit simulation supports quantum software development and verification and provides a baseline for assessing quantum advantage. We compare gate-based statevector simulation, direct Pauli evolution and matrix product states for molecular and spin Variational Hamiltonian Ansatze and hardware-efficient ansatze on CPU/GPU systems, varying width, depth, precision and resource allocation. In the tested implementations, direct Pauli evolution benefits complex molecular VHA, whereas GPUs accelerate wider spin and hardware efficient circuits. Matrix product state is useful when limited entanglement permits accurate compression; corrected CPU FP32 can meet the tested tolerance while costing more than FP64. Workload size, startup, worker memory and task-specific profiling can change the preferred allocation. Matched NumPy Pauli updates accelerate two-layer full LiH by 3.45-fold; the GPU gives approximately 30-fold acceleration over 18-thread Aer for 28-qubit, four-layer TFIM, both in FP64. We provide a procedure for choosing representation, precision and worker allocation jointly, comparing the cost of complete energy evaluations under accuracy, memory and core constraints.

## Tackling GPU Bottlenecks in Plane-Wave DFT with K-Point Batching and Mixed Precision

DevFeed: [Tackling GPU Bottlenecks in Plane-Wave DFT with K-Point Batching and Mixed Precision](<https://devfeed.tech/articles/tackling-gpu-bottlenecks-in-plane-wave-dft-with-k-point-batching-and-mixed-precision-141012.md>)

Original publisher: [Read original article](<https://developer.nvidia.com/blog/tackling-gpu-bottlenecks-in-plane-wave-dft-with-k-point-batching-and-mixed-precision/>)

Author: Erica Tsai

Published: 2026-11-18T15:00:00Z

Content type: article

Language: en

Sources: [NVIDIA Developer](<https://devfeed.tech/sources/nvidia-developer.md>), [NVIDIA Technical Blog](<https://devfeed.tech/sources/nvidia-technical-blog.md>)

Topics: [Tensor Cores](<https://devfeed.tech/topics/tensor-cores.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [aimd](<https://devfeed.tech/tags/aimd.md>), [batching](<https://devfeed.tech/tags/batching.md>), [bottlenecks](<https://devfeed.tech/tags/bottlenecks.md>), [cloud-services](<https://devfeed.tech/tags/cloud-services.md>), [computational-chemistry-materials-science](<https://devfeed.tech/tags/computational-chemistry-materials-science.md>), [developer-tools-techniques](<https://devfeed.tech/tags/developer-tools-techniques.md>), [dft](<https://devfeed.tech/tags/dft.md>), [edge-computing](<https://devfeed.tech/tags/edge-computing.md>), [energy](<https://devfeed.tech/tags/energy.md>), [featured](<https://devfeed.tech/tags/featured.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [learning](<https://devfeed.tech/tags/learning.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [parallel-programming](<https://devfeed.tech/tags/parallel-programming.md>), [precision](<https://devfeed.tech/tags/precision.md>), [quantum-chemistry](<https://devfeed.tech/tags/quantum-chemistry.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [simulation-modeling-design](<https://devfeed.tech/tags/simulation-modeling-design.md>), [software-defined-data-center](<https://devfeed.tech/tags/software-defined-data-center.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

Daviator targets underused GPUs in plane-wave DFT by batching independent k-point and spin work, including the Davidson solver, to reduce kernel launches and raise GPU occupancy. It integrates at VASP call boundaries; users enable it with LDAVIATOR=True, and current use is limited to a single node. A mixed-precision option runs less precision-sensitive work in FP32 while retaining FP64 for sensitive steps, switching back as convergence requires. The gains depend on hardware: mixed precision offers only marginal end-to-end improvement on H100/H200, but about 2x on GPUs with high FP32:FP64 throughput such as RTX PRO 6000. Benchmarks report up to 22x faster execution and 60-75% lower projected dataset-generation cost versus CPU, depending on configuration. These estimates use 13 representative OMat24 inputs and tuned parallel runs; they do not establish identical gains for every workload. A VASP user license is required.

### Source excerpt

Training a next-generation machine learning interatomic potential (MLIP) requires ground truth: tens to hundreds of millions of density functional theory (DFT)...

## Research acceleration: The jagged frontier

DevFeed: [Research acceleration: The jagged frontier](<https://devfeed.tech/articles/research-acceleration-the-jagged-frontier-142333.md>)

Original publisher: [Read original article](<https://openai.com/index/research-acceleration-jagged-frontier>)

Published: 2026-11-18T08:00:00Z

Content type: news

Language: en

Sources: [OpenAI News](<https://devfeed.tech/sources/openai-news.md>)

Topics: [AI Research](<https://devfeed.tech/topics/ai-research.md>), [ai safety](<https://devfeed.tech/topics/ai-safety.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-development](<https://devfeed.tech/tags/ai-development.md>), [ai-research](<https://devfeed.tech/tags/ai-research.md>), [ai-research-and-development](<https://devfeed.tech/tags/ai-research-and-development.md>), [alignment](<https://devfeed.tech/tags/alignment.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [openai](<https://devfeed.tech/tags/openai.md>), [recursive-self-improvement](<https://devfeed.tech/tags/recursive-self-improvement.md>), [research](<https://devfeed.tech/tags/research.md>)

### AI overview

OpenAI reports that coding agents are being used much more intensively in research, but their contribution remains uneven and closely supervised. Estimated autonomous success on tasks requiring two to four hours rose from 23% in March to 46% in October; success varied by task, from 79% for monitoring training runs to 25% for writing inference and serving code. Researchers reported substantial value from agents, while interviews described weak research judgment, quality problems, and the need for human review. Safety and Alignment documented workweeks grew 77% from January to September. Research activity measures also rose, including experiments per active experimenter reaching 1.68 times the 2025 baseline over the four weeks ending October 18. OpenAI's internal capabilities index frontier rose 16.4 points annually since August 2025, though the report cautions that activity metrics do not necessarily mean faster progress and historical trends may not predict future AI-driven gains. The findings are partial and specific to OpenAI's research organization.

### Source excerpt

OpenAI measures how coding agents are accelerating AI research, where their performance remains uneven, and what this signals about recursive self-improvement.

## TAILROLL: Tail-Aware Rollout Acceleration for RL-Based LLM Post-Training

DevFeed: [TAILROLL: Tail-Aware Rollout Acceleration for RL-Based LLM Post-Training](<https://devfeed.tech/articles/tailroll-tail-aware-rollout-acceleration-for-rl-based-llm-post-training-140384.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22206>)

Author: Niloofar Gholipour, Marcos Assuncao, Timothy Yu, Gursimran Singh, Rajkumar Buyya, Julien Gascon-Samson, Zhenan Fan, Yong Zhang

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [grpo](<https://devfeed.tech/topics/grpo.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [LLM Techniques](<https://devfeed.tech/topics/llm-techniques.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [asynchronous](<https://devfeed.tech/tags/asynchronous.md>), [balancing](<https://devfeed.tech/tags/balancing.md>), [bubble](<https://devfeed.tech/tags/bubble.md>), [cache](<https://devfeed.tech/tags/cache.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [decoding](<https://devfeed.tech/tags/decoding.md>), [engine](<https://devfeed.tech/tags/engine.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [kv-cache](<https://devfeed.tech/tags/kv-cache.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [latency](<https://devfeed.tech/tags/latency.md>), [llm](<https://devfeed.tech/tags/llm.md>), [masking](<https://devfeed.tech/tags/masking.md>), [migration](<https://devfeed.tech/tags/migration.md>), [models](<https://devfeed.tech/tags/models.md>), [online](<https://devfeed.tech/tags/online.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [rl](<https://devfeed.tech/tags/rl.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

TAILROLL targets two rollout bottlenecks in RL post-training: uneven request placement and waiting for the longest response. Its online scheduler migrates decoding requests between engines at chunk boundaries, resuming them with one re-prefill, and can terminate a response once its GRPO group has enough completed samples for advantage estimation; partial output is masked from training loss. Across ten configurations, it reports 21-74% lower rollout bubble ratio and 44-79% lower cross-engine KV-cache imbalance. End-to-end results include 24-34% less rollout time, 21-22% shorter full steps on long-CoT RL, 28-37% lower p95 latency, and up to 33% higher generation throughput, with comparable or higher final reward.

### Source excerpt

arXiv:2611.22206v1 Announce Type: new Abstract: RL post-training of large language models (LLMs) is bottlenecked by rollout: heavy-tailed responses leave engines idle as each step waits for the longest response. Asynchronous training masks, but does not eliminate, idle time. We distinguish imbalance waste (uneven placement and decode contention) from straggler waste (waiting for the longest response). Tail length, not average load, often sets the step bound. We propose TAILROLL, an online scheduler integrated into an open-source RL framework. It combines stateless in-flight migration and sufficiency-gated termination. Migration moves decoding requests between engines at chunk boundaries and resumes them with one re-prefill under a freeness-based, cost-aware tail-balancing policy. Termination stops a response when its GRPO group has enough completed samples for advantage estimation, masking its partial output from the training loss. Across ten configurations, TAILROLL reduces rollout bubble ratio by 21% to 74% and cross-engine KV-cache imbalance by 44% to 79%. In end-to-end RL training, it cuts rollout time by 24% to 34%, full step time by 21% to 22% on long-CoT RL, and p95 request latency by 28% to 37%, while raising generation throughput by up to 33% with comparable or higher final reward.

## Transferable SCF-Acceleration through Solver-Aligned Initialization Learning

DevFeed: [Transferable SCF-Acceleration through Solver-Aligned Initialization Learning](<https://devfeed.tech/articles/transferable-scf-acceleration-through-solver-aligned-initialization-learning-141550.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2604.21657>)

Author: Eike S. Eberhard, Viktor Kotsev, Timm G\\"uthle, Stephan G\\"unnemann

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Post-training optimization](<https://devfeed.tech/topics/post-training-optimization.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [convergence](<https://devfeed.tech/tags/convergence.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dft](<https://devfeed.tech/tags/dft.md>), [extrapolation](<https://devfeed.tech/tags/extrapolation.md>), [hamiltonian](<https://devfeed.tech/tags/hamiltonian.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [iteration](<https://devfeed.tech/tags/iteration.md>), [large](<https://devfeed.tech/tags/large.md>), [learning](<https://devfeed.tech/tags/learning.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [matrix](<https://devfeed.tech/tags/matrix.md>), [models](<https://devfeed.tech/tags/models.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [parts](<https://devfeed.tech/tags/parts.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [quality](<https://devfeed.tech/tags/quality.md>), [reduce](<https://devfeed.tech/tags/reduce.md>), [reduction](<https://devfeed.tech/tags/reduction.md>), [required](<https://devfeed.tech/tags/required.md>), [sail](<https://devfeed.tech/tags/sail.md>)

### AI overview

Ground-state supervision can yield accurate out-of-distribution predictions that nevertheless slow SCF convergence. Solver-Aligned Initialization Learning (SAIL) addresses this by differentiating through parts of the SCF solver when training Hamiltonian and density matrix models. The authors also introduce ERIC, which adjusts the commonly used RIC to account for hidden Fock-build overhead. On QM40, with molecules up to four times larger than the training distribution, SAIL reduces ERIC by 40% for PBE, 35% for SCAN, and 32% for B3LYP; on QMugs molecules ten times larger than training, it reports a 1.35x wall-time speedup at the hybrid level of theory. These findings are reported in the supplied abstract.

### Source excerpt

arXiv:2604.21657v3 Announce Type: replace Abstract: The cost of Kohn-Sham density functional theory (KS-DFT) calculations scales with the number of required solver iterations, which depend on the quality of the initial guess. Machine learning methods that predict initial guesses from molecular geometry can reduce this cost, but matrix-prediction models fail when extrapolating to larger molecules, degrading rather than accelerating convergence [Liu et al., 2025]. We show that this failure is a supervision problem, not an extrapolation problem: models trained on ground-state targets fit those targets well out of distribution, yet produce initial guesses that slow convergence. Solver-Aligned Initialization Learning (SAIL) resolves this for both Hamiltonian and density matrix models by differentiating through parts of the self-consistent field (SCF) solver end-to-end. We introduce the Effective Relative Iteration Count (ERIC), a correction to the commonly used RIC that accounts for hidden Fock-build overhead. On QM40, which contains molecules up to 4$\times$ larger than the training distribution, SAIL reduces ERIC by 40% (PBE), 35% (SCAN), and 32% (B3LYP), more than doubling the previous state-of-the-art reduction on B3LYP. On QMugs molecules 10$\times$ larger than the training set, SAIL delivers a 1.35$\times$ wall-time speedup at the hybrid level of theory, extending ML SCF acceleration to large drug-like molecules.

## Android UI thread vs. RenderThread: what actually happens when a frame is drawn

DevFeed: [Android UI thread vs. RenderThread: what actually happens when a frame is drawn](<https://devfeed.tech/articles/android-ui-thread-vs-renderthread-what-actually-happens-when-a-frame-is-drawn-140454.md>)

Original publisher: [Read original article](<https://proandroiddev.com/android-ui-thread-vs-renderthread-3f54bb74340f?source=rss----c72404660798---4>)

Author: Ravi Rawal

Published: 2026-11-17T19:50:28Z

Content type: article

Language: en

Sources: [ProAndroidDev](<https://devfeed.tech/sources/proandroiddev-medium.md>)

Topics: [Android](<https://devfeed.tech/topics/android.md>), [jank](<https://devfeed.tech/topics/jank.md>), [real-time rendering](<https://devfeed.tech/topics/real-time-rendering.md>), [Jetpack](<https://devfeed.tech/topics/jetpack.md>)

Tags: [accelerated](<https://devfeed.tech/tags/accelerated.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [also](<https://devfeed.tech/tags/also.md>), [android](<https://devfeed.tech/tags/android.md>), [android-app-development](<https://devfeed.tech/tags/android-app-development.md>), [animation](<https://devfeed.tech/tags/animation.md>), [app](<https://devfeed.tech/tags/app.md>), [application](<https://devfeed.tech/tags/application.md>), [apps](<https://devfeed.tech/tags/apps.md>), [background](<https://devfeed.tech/tags/background.md>), [blocking](<https://devfeed.tech/tags/blocking.md>), [buffer](<https://devfeed.tech/tags/buffer.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [canvas](<https://devfeed.tech/tags/canvas.md>), [cause](<https://devfeed.tech/tags/cause.md>), [changes](<https://devfeed.tech/tags/changes.md>), [frame](<https://devfeed.tech/tags/frame.md>), [kotlin](<https://devfeed.tech/tags/kotlin.md>), [performance](<https://devfeed.tech/tags/performance.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [thread](<https://devfeed.tech/tags/thread.md>), [ui](<https://devfeed.tech/tags/ui.md>), [vs](<https://devfeed.tech/tags/vs.md>)

### AI overview

Frame drops can persist even when app code is off the main thread because frame preparation, command rendering, GPU execution, and display composition are separate stages. The UI thread handles callbacks, layout and drawing; with hardware acceleration, onDraw() records commands into display lists rather than rasterizing them. RenderThread replays those lists and submits graphics work, the GPU executes it, and SurfaceFlinger composes the app buffer. Diagnose the slow stage with a system trace: inspect main-thread Choreographer work, RenderThread DrawFrame, GPU completion, and presentation timing. Heavy onDraw() calculations can delay submission; costly paths or clips can make RenderThread slow; textures, overdraw, or shaders can saturate the GPU. Cache geometry or simplify commands according to the bottleneck. Compose uses the same graphics pipeline, though state reads can be placed in different stages to limit which work reruns. The supplied text is an excerpt and does not establish full-document coverage.

### Source excerpt

An Android app with a non-blocking main thread in your application can still get frame drops. This seems odd as we know that drawing a frame involves several components. Our code will only run in the first component. The UI thread will handle the app-side of things and records the drawing commands. A Native thread (RenderThread) will take these recordings and do the actual rendering. The GPU will then execute the graphics commands and SurfaceFlinger will compose your apps buffer with everything else on screen. This can be confusing when trying to solve dropped frames. Slowness in your onDraw() and an overloaded GPU will look the same to your users, but need different solutions. Moving your code to a background thread will only help with one of these issues. To find out which issue you have, follow your frame through the pipeline. What the UI thread does Every app has a main thread, usually called the UI thread. It owns the main message queue and runs lifecycle callbacks, input events and anything posted to the main looper. For the View system it also applies property changes, measures and lays out the hierarchy, runs animation callbacks, calls View.draw() and your onDraw(), and records drawing operations for hardware-accelerated rendering. The last job is easy to overlook. With hardware acceleration, onDraw() does not rasterize anything. Your Canvas calls are recorded into a display list, and each View keeps its own (a RenderNode). Here is a simple custom View: class ProgressView(context: Context) : View(context) { private val paint = Paint(Paint.ANTI_ALIAS_FLAG) private var progress = 0.5f override fun onDraw(canvas: Canvas) { super.onDraw(canvas) canvas.drawCircle(width / 2f, height / 2f, width * progress / 2f, paint) } } When invalidate() schedules a redraw, the framework calls onDraw() during the next frame traversal, on the UI thread, and the circle is recorded rather than painted. That is why a custom View can cause jank while the GPU is almost idle. If onDraw

## Accelerating development for massive Git repositories

DevFeed: [Accelerating development for massive Git repositories](<https://devfeed.tech/articles/accelerating-development-for-massive-git-repositories-138168.md>)

Original publisher: [Read original article](<https://developers.googleblog.com/accelerating-development-for-massive-git-repositories/>)

Author: The Android ABFS Team

Published: 2026-11-17T18:09:46.914988Z

Content type: article

Language: en

Sources: [Google Developers Blog](<https://devfeed.tech/sources/google-developers-blog.md>)

Topics: [aosp](<https://devfeed.tech/topics/aosp.md>), [Git](<https://devfeed.tech/topics/git.md>), [bazel](<https://devfeed.tech/topics/bazel.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Gerrit](<https://devfeed.tech/topics/gerrit.md>)

Tags: [17](<https://devfeed.tech/tags/17.md>), [2](<https://devfeed.tech/tags/2.md>), [3](<https://devfeed.tech/tags/3.md>), [accelerated](<https://devfeed.tech/tags/accelerated.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [android](<https://devfeed.tech/tags/android.md>), [build](<https://devfeed.tech/tags/build.md>), [builds](<https://devfeed.tech/tags/builds.md>), [cache](<https://devfeed.tech/tags/cache.md>), [content](<https://devfeed.tech/tags/content.md>), [development](<https://devfeed.tech/tags/development.md>), [file](<https://devfeed.tech/tags/file.md>), [filesystem](<https://devfeed.tech/tags/filesystem.md>), [git](<https://devfeed.tech/tags/git.md>), [google](<https://devfeed.tech/tags/google.md>), [open](<https://devfeed.tech/tags/open.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [repositories](<https://devfeed.tech/tags/repositories.md>), [service](<https://devfeed.tech/tags/service.md>), [source](<https://devfeed.tech/tags/source.md>)

### AI overview

ABFS speeds work on large Android repositories by mounting a virtual filesystem that fetches file contents on demand and shares cached build results. Its Linux kernel module, host daemons, and server use content hashes to store Git objects, workspace changes, and build outputs. CASFS exposes file metadata before downloading bytes; its tap utility records files a build reads or changes, and Ninja-go uses content-addressable caching for Ninja builds. The article reports that agents can mount Android release branches in under eight seconds, check out a tree in under ten seconds, and complete a full build in under ten minutes. These results are from runs on a 64-vCPU GCE VM with the ABFS server in the same cloud network. ABFS is open source; Android Soong/Ninja was its first target, with a Bazel proof of concept and other projects exploring compatibility.

### Source excerpt

Meet Another Build File Service (ABFS), Google's open-source content-addressable filesystem and build cache that cuts Android checkouts to <8s and builds to ~2m.

## ApkStart runs Android apps on Windows like regular desktop software

DevFeed: [ApkStart runs Android apps on Windows like regular desktop software](<https://devfeed.tech/articles/apkstart-runs-android-apps-on-windows-like-regular-desktop-software-137902.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/17/github-routinemiku-apkstart-windows/>)

Author: Anne Barela

Published: 2026-11-17T16:07:18Z

Content type: article

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [Windows](<https://devfeed.tech/topics/windows.md>), [Android](<https://devfeed.tech/topics/android.md>), [APK](<https://devfeed.tech/topics/apk.md>), [adb](<https://devfeed.tech/topics/adb.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [adb](<https://devfeed.tech/tags/adb.md>), [android](<https://devfeed.tech/tags/android.md>), [android-apps](<https://devfeed.tech/tags/android-apps.md>), [android-studio](<https://devfeed.tech/tags/android-studio.md>), [apk](<https://devfeed.tech/tags/apk.md>), [computers](<https://devfeed.tech/tags/computers.md>), [desktop](<https://devfeed.tech/tags/desktop.md>), [driver](<https://devfeed.tech/tags/driver.md>), [emulator](<https://devfeed.tech/tags/emulator.md>), [gaming](<https://devfeed.tech/tags/gaming.md>), [github](<https://devfeed.tech/tags/github.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [hardware-acceleration](<https://devfeed.tech/tags/hardware-acceleration.md>), [ico](<https://devfeed.tech/tags/ico.md>), [offline](<https://devfeed.tech/tags/offline.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [orientation](<https://devfeed.tech/tags/orientation.md>), [qemu](<https://devfeed.tech/tags/qemu.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [software](<https://devfeed.tech/tags/software.md>), [window](<https://devfeed.tech/tags/window.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

ApkStart runs Android apps on Windows in separate windows with desktop icons extracted from their APKs. Double-clicking an APK installs it; ApkStart downloads, checks, and stores its QEMU, Android emulator, and system image runtime. It reads AndroidManifest.xml offline to determine an app's orientation and saves that setting in apps.json. Its adb server uses port 5044 to avoid conflicts with Android Studio or scrcpy, and it supports WHPX or AEHD hardware acceleration. The installer is about 1.32GB. The code is MIT licensed, while release packages omit Android system images and Google components.

### Source excerpt

ApkStart, from developer RoutineMiku and published on GitHub, runs Android apps on Windows like regular desktop software. Double-clicking an APK installs it, and the app then opens in its own window with its own desktop icon, pulled out of the APK as an .ico file. The project manages its own runtime: QEMU, the Android emulator [...]

## Nvidia reveals Rubin HPC, a GPU with 50,000+ CUDA cores for scientific simulation

DevFeed: [Nvidia reveals Rubin HPC, a GPU with 50,000+ CUDA cores for scientific simulation](<https://devfeed.tech/articles/nvidia-reveals-rubin-hpc-a-gpu-with-50-000-cuda-cores-for-scientific-simulation-137868.md>)

Original publisher: [Read original article](<https://www.techspot.com/news/114583-nvidia-reveals-rubin-hpc-gpu-50000-cuda-cores.html>)

Author: Skye Jacobs

Published: 2026-11-17T15:46:00Z

Content type: news

Language: en

Sources: [TechSpot](<https://devfeed.tech/sources/techspot.md>)

Topics: [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [CUDA-X](<https://devfeed.tech/topics/cuda-x.md>), [NVIDIA Vera Rubin](<https://devfeed.tech/topics/nvidia-vera-rubin.md>), [CUDA](<https://devfeed.tech/topics/cuda.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>)

Tags: [accelerated-computing](<https://devfeed.tech/tags/accelerated-computing.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-workflows](<https://devfeed.tech/tags/agentic-workflows.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [amd](<https://devfeed.tech/tags/amd.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [multi-core](<https://devfeed.tech/tags/multi-core.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [rubin](<https://devfeed.tech/tags/rubin.md>), [rubin-gpu](<https://devfeed.tech/tags/rubin-gpu.md>), [rubin-hpc](<https://devfeed.tech/tags/rubin-hpc.md>), [science](<https://devfeed.tech/tags/science.md>), [simulation](<https://devfeed.tech/tags/simulation.md>)

### AI overview

Nvidia says its Rubin HPC GPU, announced for the Vera Rubin platform and expected in 2028, will have more than 50,000 CUDA cores and target scientific and engineering simulations used to test ideas generated by AI agents. The company says its "balanced design" scales data movement with compute, using sparse data acceleration for recurring routines such as sparse matrix operations, stencil computations, sorting and reductions. GPUs will connect through seventh-generation NVLink, and agents can use CUDA-X libraries with tuned algorithms. Key HPC specifications remain undisclosed: memory capacity and bandwidth, double-precision capability, interconnect bandwidth, and whether the chip uses multiple dies. Nvidia has not confirmed its physical layout.

### Source excerpt

The core count offers the first clue to what Nvidia is building. The current Rubin GPU has 224 streaming multiprocessors and 28,672 CUDA cores, so getting past 50,000 would take at least 74% more cores. Read Entire Article

## Value and Policy Calibration from Execution Feedback for Socially Compatible Driving

DevFeed: [Value and Policy Calibration from Execution Feedback for Socially Compatible Driving](<https://devfeed.tech/articles/value-and-policy-calibration-from-execution-feedback-for-socially-compatible-driving-137516.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19969>)

Author: Sen Shen, Ning Liu, Xiangrui Kong, Yingwei Li, Dongkun Han

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [autonomous vehicles](<https://devfeed.tech/topics/autonomous-vehicles.md>), [World models](<https://devfeed.tech/topics/world-models.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [actor](<https://devfeed.tech/tags/actor.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-vehicle](<https://devfeed.tech/tags/autonomous-vehicle.md>), [av](<https://devfeed.tech/tags/av.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [delay](<https://devfeed.tech/tags/delay.md>), [driving](<https://devfeed.tech/tags/driving.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [framework](<https://devfeed.tech/tags/framework.md>), [imagination](<https://devfeed.tech/tags/imagination.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [intersection](<https://devfeed.tech/tags/intersection.md>), [learning](<https://devfeed.tech/tags/learning.md>), [model](<https://devfeed.tech/tags/model.md>), [policy](<https://devfeed.tech/tags/policy.md>), [value](<https://devfeed.tech/tags/value.md>)

### AI overview

The proposed world-model reinforcement learning framework combines imagined policy improvement with calibration from finalized interactions: off-policy correction and replay updates use observed rewards and transitions to supervise both the critic and actor. In simulation at an unsignalized intersection, it achieved 100% success and reduced following-HDV acceleration variation by 65.5% versus PIWM and 44.0% versus DreamerV3. Ablations attribute reduced HDV burden to value calibration and improved passage efficiency to policy calibration. Under out-of-distribution traffic shifts, it reported lower ego jerk and HDV acceleration variation than PIWM at comparable passage times. These are simulation results; the supplied text does not establish real-world performance.

### Source excerpt

arXiv:2611.19969v1 Announce Type: cross Abstract: Autonomous vehicle (AV) decisions at intersections affect the delay, braking, and motion smoothness of surrounding human-driven vehicles (HDVs). Learning socially compatible behavior requires connecting imagined interaction outcomes with feedback from executed actions. We propose a world-model reinforcement learning framework combines imagined policy improvement with direct value and policy calibration from finalized interactions, balancing AV progress with HDV interaction burden. An interaction-aware latent model supports imagination, while off-policy correction and replay updates enable observed rewards and transitions to supervise both the critic and actor. Simulation experiments at an unsignalized intersection achieve a 100% success and reductions in following-HDV acceleration variation of 65.5% and 44.0% relative to PIWM and DreamerV3, respectively. Ablations identify complementary roles for value calibration in reducing HDV burden and policy calibration in improving passage efficiency. Tests under out-of-distribution traffic shifts further show lower ego jerk and HDV acceleration variation than PIWM at comparable passage times.

## Transferable neural-network preconditioners for Wilson--Clover propagator inversions

DevFeed: [Transferable neural-network preconditioners for Wilson--Clover propagator inversions](<https://devfeed.tech/articles/transferable-neural-network-preconditioners-for-wilson-clover-propagator-inversions-139118.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20990>)

Author: Shiyi Zhong, Yuanyi Zhang, Jian-Hui Zhang, Dian-Jun Zhao

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [cnn](<https://devfeed.tech/tags/cnn.md>), [combined](<https://devfeed.tech/tags/combined.md>), [convolution](<https://devfeed.tech/tags/convolution.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [gauge](<https://devfeed.tech/tags/gauge.md>), [hep-lat](<https://devfeed.tech/tags/hep-lat.md>), [lattice](<https://devfeed.tech/tags/lattice.md>), [layer](<https://devfeed.tech/tags/layer.md>), [network](<https://devfeed.tech/tags/network.md>), [neural-network](<https://devfeed.tech/tags/neural-network.md>), [physics-comp-ph](<https://devfeed.tech/tags/physics-comp-ph.md>), [ptc](<https://devfeed.tech/tags/ptc.md>), [reduce](<https://devfeed.tech/tags/reduce.md>), [scheme](<https://devfeed.tech/tags/scheme.md>), [small](<https://devfeed.tech/tags/small.md>), [volume](<https://devfeed.tech/tags/volume.md>)

### AI overview

Two single-layer neural-network approaches accelerate four-dimensional Wilson-Clover propagator inversions and transfer to new ensembles without retraining. A gauge-covariant parallel transport convolution (PTC) network directly reduces generalized conjugate residual (GCR) iterations and maintains comparable acceleration across lattice volumes, lattice spacings, and pion masses. A CNN-like network paired with a multigrid-inspired algorithm transfers from small to large volumes with strong acceleration. Combining the approaches further reduces GCR iterations. A cost comparison that includes basis construction and solves finds the scheme more advantageous for large-volume than small-volume ensembles.

### Source excerpt

arXiv:2611.20990v1 Announce Type: cross Abstract: We investigate two single--layer neural-network architectures for accelerating four-dimensional Wilson--Clover propagator inversions on lattice ensembles in practical use; both transfer directly without retraining. The first is a gauge-covariant parallel transport convolution (PTC) network that directly reduces the number of generalized conjugate residual (GCR) solver iterations. The second is a CNN-like network used together with a multigrid-inspired algorithm. The two approaches can be combined to further reduce GCR iterations. PTC transfers without retraining across lattice volumes, lattice spacings, and pion masses while maintaining comparable acceleration, whereas the CNN-like network transfers directly from small- to large-volume ensembles with strong acceleration. A cost comparison accounting for both basis construction and solves shows that the proposed scheme is more advantageous for large-volume ensembles than for small-volume ones.

## artist: A Python package for differentiable ray tracing and gradient-based optimization of solar tower power plant operation

DevFeed: [artist: A Python package for differentiable ray tracing and gradient-based optimization of solar tower power plant operation](<https://devfeed.tech/articles/artist-a-python-package-for-differentiable-ray-tracing-and-gradient-based-optimization-of-solar-tower-power-plant-operation-139101.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20851>)

Author: Marlene Busch (Institute of Solar Research, German Aerospace Center), Marie Weiel (Scientific Computing Center, Helmholtz AI, Karlsruhe, Germany), Kaleb Phipps (Scientific Computing Center, Helmholtz

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Ray Tracing](<https://devfeed.tech/topics/ray-tracing.md>), [Ray Tracing / Path Tracing](<https://devfeed.tech/topics/ray-tracing-path-tracing.md>), [MLX](<https://devfeed.tech/topics/mlx.md>), [Content Creation / Rendering](<https://devfeed.tech/topics/content-creation-rendering.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [blocking](<https://devfeed.tech/tags/blocking.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [database](<https://devfeed.tech/tags/database.md>), [digital](<https://devfeed.tech/tags/digital.md>), [digital-twin](<https://devfeed.tech/tags/digital-twin.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [package](<https://devfeed.tech/tags/package.md>), [plant](<https://devfeed.tech/tags/plant.md>), [power](<https://devfeed.tech/tags/power.md>), [power-plant](<https://devfeed.tech/tags/power-plant.md>), [python](<https://devfeed.tech/tags/python.md>), [ray](<https://devfeed.tech/tags/ray.md>), [ray-tracing](<https://devfeed.tech/tags/ray-tracing.md>), [solar](<https://devfeed.tech/tags/solar.md>), [tower](<https://devfeed.tech/tags/tower.md>), [tracing](<https://devfeed.tech/tags/tracing.md>)

### AI overview

artist is an open-source Python package that makes solar tower optical modeling differentiable end to end, allowing gradient-based inference and optimization beyond conventional ray tracers' forward-only simulations. Its PyTorch digital twin models heliostat surfaces and kinematics, blocking, and ray tracing. The abstract says it can reconstruct surfaces and kinematics from operational flux images without costly deflectometry, predict receiver irradiance, and optimize aim points for a target flux distribution. GPU acceleration, a structure-of-arrays layout, nested data parallelism, and links to open data such as PAINT are presented as ways to support real-plant workflows. The supplied abstract describes representative demonstrations but gives no quantitative performance results.

### Source excerpt

arXiv:2611.20851v1 Announce Type: new Abstract: Solar tower power plants provide dispatchable, renewable energy by concentrating sunlight with thousands of mirrors, called heliostats, onto a central receiver. Efficient operation requires accurate modeling of the optical path from each heliostat to the receiver and compensation for manufacturing imperfections, kinematic misalignment, and material degradation that distort the flux distribution. Conventional ray tracers predict the flux from an assumed plant model but are limited to forward simulation. They cannot recover unknown plant parameters from measured flux nor directly optimize plant operation. We present artist, an open-source, FAIR4RS-compliant Python package implementing a fully differentiable digital twin of solar tower power plants. By making the optical pipeline, comprising heliostat surface representation, kinematics, blocking, and ray tracing, end to end differentiable, artist turns the digital twin from a passive forward model into an active physics-based tool for inverse modeling and data-driven optimization. The resulting gradient-based framework unifies tasks usually handled by separate solutions, such as reconstructing heliostat surfaces and kinematics from operational flux images without costly deflectometry, predicting irradiance at the receiver, and optimizing aim points to realize a desired receiver flux density distribution. Built on PyTorch, artist scales to real plants through GPU acceleration, a structure-of-arrays memory layout, and nested data parallelism. It interfaces with open data sources such as the PAINT database. We describe the software architecture, physical models, and optimization framework, and demonstrate the workflow on representative tasks. By integrating modeling, prediction, and optimization into a single open, interpretable package, artist provides a principled foundation for data-driven research toward autonomous solar tower operation.

## IBM, AMD advance quantum error correction decoding at scale

DevFeed: [IBM, AMD advance quantum error correction decoding at scale](<https://devfeed.tech/articles/ibm-amd-advance-quantum-error-correction-decoding-at-scale-136373.md>)

Original publisher: [Read original article](<https://research.ibm.com/blog/pelagos>)

Author: Blake Johnson; Robert Davis

Published: 2026-11-16T15:59:00Z

Content type: news

Language: en

Sources: [IBM Research](<https://devfeed.tech/sources/ibm-research.md>)

Topics: [Quantum Computing](<https://devfeed.tech/topics/quantum-computing.md>), [Decoding](<https://devfeed.tech/topics/decoding.md>), [rocm](<https://devfeed.tech/topics/rocm.md>), [ibm](<https://devfeed.tech/topics/ibm.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [accelerators](<https://devfeed.tech/tags/accelerators.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [amd](<https://devfeed.tech/tags/amd.md>), [decoding](<https://devfeed.tech/tags/decoding.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [ibm](<https://devfeed.tech/tags/ibm.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [quantum-error-correction](<https://devfeed.tech/tags/quantum-error-correction.md>), [quantum-research](<https://devfeed.tech/tags/quantum-research.md>), [quantum-software](<https://devfeed.tech/tags/quantum-software.md>), [release](<https://devfeed.tech/tags/release.md>), [scale](<https://devfeed.tech/tags/scale.md>)

### AI overview

IBM and AMD released Pelagos, an open-source framework for evaluating quantum error-correction decoders at scale. Built with Triton, it supports AMD and NVIDIA GPUs and combines GPU acceleration with a reusable software architecture. Multi-GPU work uses process-level orchestration, parallel host-side operations and shared memory to reduce coordination overhead. Tests on AMD Instinct MI300X and MI355X showed up to roughly 7.5x speedup in the GPU decoding stage across eight GPUs versus one; the team also deployed Pelagos on ORNL's Lux HPC system. The excerpt says earlier studies found up to 97x higher throughput than a single CPU core. An accompanying survey reviews decoding methods and platforms including CPUs, GPUs, FPGAs and specialized accelerators, and argues that classical infrastructure and hardware-software co-design are key parts of building fault-tolerant quantum systems.

### Source excerpt

New open-source tool from IBM and AMD helps researchers evaluate quantum error-correction decoders at scale for future fault-tolerant quantum systems.

## AMD Unveils Versal TR Devices With 229,000 DMIPS and Backside Attack Protection

DevFeed: [AMD Unveils Versal TR Devices With 229,000 DMIPS and Backside Attack Protection](<https://devfeed.tech/articles/amd-unveils-versal-tr-devices-with-229-000-dmips-and-backside-attack-protection-136076.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/amd-unveils-versal-tr-devices-with-229000-dmips-and-backside-attack-protection>)

Author: Harold Fritts

Published: 2026-11-16T15:43:01Z

Content type: news

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [Protection](<https://devfeed.tech/topics/protection.md>), [Embedded Systems](<https://devfeed.tech/topics/embedded-systems.md>), [Secrets Management](<https://devfeed.tech/topics/secrets-management.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [aging](<https://devfeed.tech/tags/aging.md>), [ai](<https://devfeed.tech/tags/ai.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [amd](<https://devfeed.tech/tags/amd.md>), [application-security](<https://devfeed.tech/tags/application-security.md>), [applications](<https://devfeed.tech/tags/applications.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arm](<https://devfeed.tech/tags/arm.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [availability](<https://devfeed.tech/tags/availability.md>), [boot](<https://devfeed.tech/tags/boot.md>), [built-in](<https://devfeed.tech/tags/built-in.md>), [compute](<https://devfeed.tech/tags/compute.md>), [edge](<https://devfeed.tech/tags/edge.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [int8](<https://devfeed.tech/tags/int8.md>), [portfolio](<https://devfeed.tech/tags/portfolio.md>), [protection](<https://devfeed.tech/tags/protection.md>)

### AI overview

AMD's Versal TR devices pair Versal AI Edge Series Gen 2 compute with added physical and cryptographic protections. A sensor-based die bonded over the compute die is intended to shield the silicon backside, detect tampering, and prevent boot if it is removed or compromised; expanded glitch detection covers simultaneous glitches across two voltage rails. The devices add hardware acceleration for ML-DSA and SLH-DSA, hybrid boot signatures, and ASU key-vault improvements alongside a memory element aging sensor. AMD lists portfolio maxima of 229,000 DMIPS, 369 INT8 TOPS, and 543,000 LUTs, without assigning them to specific SKUs. Select equivalent Gen 2 devices can be used for development and transitioned to corresponding TR devices without rerunning synthesis or place and route. First samples are expected in the second half of 2027, with production in the first half of 2028. The first-of-its-kind protection claim is attributed to AMD's internal analysis; the referenced footnote was not supplied.

### Source excerpt

AMD has announced Versal tamper-resistant (TR) devices that combine physical attack protection with the heterogeneous compute architecture of the AMD Versal AI Edge Series Gen 2. The portfolio delivers up to 229,000 DMIPS of scalar compute, 369 INT8 TOPS, and 543,000 LUTs, according to AMD. The first devices are expected to sample in the second The post AMD Unveils Versal TR Devices With 229,000 DMIPS and Backside Attack Protection appeared first on StorageReview.com.

## Intel Diamond Rapids Xeon Gets Up to 1.6TB/s of Memory Bandwidth and 1.28GB of Cache Across 256 P-Cores

DevFeed: [Intel Diamond Rapids Xeon Gets Up to 1.6TB/s of Memory Bandwidth and 1.28GB of Cache Across 256 P-Cores](<https://devfeed.tech/articles/intel-diamond-rapids-xeon-gets-up-to-1-6tb-s-of-memory-bandwidth-and-1-28gb-of-cache-across-256-p-cores-136080.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/intel-diamond-rapids-1-6tb-s-memory-bandwidth-1-28gb-cache-gpu-d270-sc26>)

Author: Harold Fritts

Published: 2026-11-16T15:39:46Z

Content type: news

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [CPU Cache](<https://devfeed.tech/topics/cpu-cache.md>)

Tags: [18a](<https://devfeed.tech/tags/18a.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [accelerators](<https://devfeed.tech/tags/accelerators.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-frameworks](<https://devfeed.tech/tags/ai-frameworks.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [cache](<https://devfeed.tech/tags/cache.md>), [computing](<https://devfeed.tech/tags/computing.md>), [diamond-rapids](<https://devfeed.tech/tags/diamond-rapids.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [high-performance](<https://devfeed.tech/tags/high-performance.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [intel](<https://devfeed.tech/tags/intel.md>), [last-level-cache](<https://devfeed.tech/tags/last-level-cache.md>), [memory](<https://devfeed.tech/tags/memory.md>), [memory-bandwidth](<https://devfeed.tech/tags/memory-bandwidth.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [scale](<https://devfeed.tech/tags/scale.md>), [server](<https://devfeed.tech/tags/server.md>), [xeon](<https://devfeed.tech/tags/xeon.md>)

### AI overview

Intel says its planned Diamond Rapids Xeon will scale to 256 performance cores, up to 1.6TB/s of memory bandwidth and 1.28GB of last-level cache, with APX, AVX 10.2 and enhanced AMX. The bandwidth figure's memory configuration and whether it is per socket remain unspecified, and Intel gave no launch date. The article also describes the air-cooled PCIe GPU D270, with 160GB of LPDDR5X and 1.5TB/s bandwidth, plus partner designs up to 480GB. Intel's performance comparisons against NVIDIA are vendor results on pre-production hardware; they were not independently run or audited, and the financial comparison's per-card normalization is unclear. Intel also outlines continued investment in HPC software, silicon photonics and a proposed Classiq acquisition, pending regulatory approval.

### Source excerpt

Intel laid out its high-performance computing plans for SC26, with Diamond Rapids, its next-generation Xeon server architecture, at the center. Intel says the platform will scale to 256 performance cores with up to 1.6TB/s of memory bandwidth and up to 1.28GB of last-level cache. The outline came from Jeff McVeigh, vice president and general manager The post Intel Diamond Rapids Xeon Gets Up to 1.6TB/s of Memory Bandwidth and 1.28GB of Cache Across 256 P-Cores appeared first on StorageReview.com.

## Training-Free Acceleration of Multi-View 3D Transformers via Regime-Aware Attention Approximation

DevFeed: [Training-Free Acceleration of Multi-View 3D Transformers via Regime-Aware Attention Approximation](<https://devfeed.tech/articles/training-free-acceleration-of-multi-view-3d-transformers-via-regime-aware-attention-approximation-135089.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17629>)

Author: Muhammed Izzet Sezer, Sevval Belkis Dikkaya

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [3D](<https://devfeed.tech/topics/3d.md>), [Image](<https://devfeed.tech/topics/image.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>)

Tags: [3](<https://devfeed.tech/tags/3.md>), [3d](<https://devfeed.tech/tags/3d.md>), [about](<https://devfeed.tech/tags/about.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [accelerators](<https://devfeed.tech/tags/accelerators.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [approximation](<https://devfeed.tech/tags/approximation.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attention](<https://devfeed.tech/tags/attention.md>), [building](<https://devfeed.tech/tags/building.md>), [but](<https://devfeed.tech/tags/but.md>), [camera](<https://devfeed.tech/tags/camera.md>), [case](<https://devfeed.tech/tags/case.md>), [code](<https://devfeed.tech/tags/code.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [free](<https://devfeed.tech/tags/free.md>), [training](<https://devfeed.tech/tags/training.md>), [transformers](<https://devfeed.tech/tags/transformers.md>), [view](<https://devfeed.tech/tags/view.md>)

### AI overview

Multi-view 3D transformers use global attention whose cost grows quadratically with image count, and common sparse-attention accelerators can lose accuracy or provide little speedup. The paper introduces RAWATTA, a training-free approach that uses 8-bit attention and retains 16-bit precision in layers flagged by a worst-case calibration check. Its faster variant, RAWATTA-Fast, also approximates attention in some scene-averaging and first-image-anchoring layers, while preserving all keys for point-matching layers and the first image. On a 12 GB GPU with 50 images, RAWATTA runs VGGT faster than the released code, mostly due to its CUDA engine, with at most about a one-point AUC@3 pose-accuracy loss on any tested dataset; RAWATTA-Fast is slightly faster and less accurate. Both handle 200 images, while the released code runs out of memory at 100.

### Source excerpt

arXiv:2611.17629v1 Announce Type: cross Abstract: Multi-view 3D transformers such as VGGT reconstruct a scene from a set of unposed images in a single forward pass, predicting camera poses, depth and 3D points, but the cost of their global attention grows quadratically with the num- ber of images. Accelerators borrowed from large language models assume that each token attends to only a few oth- ers; on these models they lose accuracy or gain little speed. We study the global attention of three multi-view transform- ers on seven datasets and find that every layer plays one of three roles: it averages over the whole scene, matches points across many images, or anchors on the first image, whose coordinates the outputs use. Building on this anal- ysis, we propose Regime-AWare ATTention Approximation (RAWATTA), a training-free method that runs attention in 8- bit precision and keeps 16-bit precision only in the layers where a worst-case calibration check requires it. Its fast set- ting, RAWATTA-Fast, also approximates attention in a few averaging and anchoring layers, while matching layers and the first image keep all their keys. On a 12 GB GPU at 50 im- ages, RAWATTA runs VGGT faster than the released code, mostly through our CUDA engine, and loses at most about one point of pose accuracy (AUC@3) on any dataset; RAWATTA-Fast is slightly faster and less accurate. Both fit 200 images, where the released code runs out of memory at 100.

## Improving Matrix Exponential for Generative AI Flows: A Taylor-Based Approach Beyond Paterson--Stockmeyer

DevFeed: [Improving Matrix Exponential for Generative AI Flows: A Taylor-Based Approach Beyond Paterson--Stockmeyer](<https://devfeed.tech/articles/improving-matrix-exponential-for-generative-ai-flows-a-taylor-based-approach-beyond-paterson-stockmeyer-135658.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2512.20777>)

Author: Jorge Sastre, Daniel Faronbi, Jos\\'e Miguel Alonso, Peter Traver, Javier Ib\\'a\\~nez, Nuria Lloret

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [computational complexity](<https://devfeed.tech/topics/computational-complexity.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [ai](<https://devfeed.tech/tags/ai.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [applications](<https://devfeed.tech/tags/applications.md>), [approach](<https://devfeed.tech/tags/approach.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [based](<https://devfeed.tech/tags/based.md>), [combined](<https://devfeed.tech/tags/combined.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [computational-complexity](<https://devfeed.tech/tags/computational-complexity.md>), [computing](<https://devfeed.tech/tags/computing.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-na](<https://devfeed.tech/tags/cs-na.md>), [develop](<https://devfeed.tech/tags/develop.md>), [effort](<https://devfeed.tech/tags/effort.md>), [error](<https://devfeed.tech/tags/error.md>), [generative](<https://devfeed.tech/tags/generative.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [math-na](<https://devfeed.tech/tags/math-na.md>), [matrix](<https://devfeed.tech/tags/matrix.md>)

### AI overview

The paper proposes an optimized Taylor-based matrix exponential algorithm for high-throughput generative AI flows. It uses error analysis and dynamically selects Taylor order and scaling factor to minimize computation under a prescribed error tolerance. The abstract reports that numerical experiments show acceleration and high numerical stability versus existing state-of-the-art implementations, but provides no quantitative results or experiment details. Its comparison with Padé scaling-and-squaring and Paterson-Stockmeyer is broad; specific implementation or performance tradeoffs are not given.

### Source excerpt

arXiv:2512.20777v3 Announce Type: replace Abstract: The matrix exponential is a fundamental operator in scientific computing and system simulation, with applications ranging from control theory and quantum mechanics to modern generative machine learning. While Pad\'e approximants combined with scaling and squaring have long served as the standard, recent Taylor-based methods, which utilize polynomial evaluation schemes that surpass the classical Paterson--Stockmeyer technique, offer superior accuracy and reduced computational complexity. This paper presents an optimized Taylor-based algorithm for the matrix exponential, specifically designed for the high-throughput requirements of generative AI flows. We provide a rigorous error analysis and develop a dynamic selection strategy for the Taylor order and scaling factor to minimize computational effort under a prescribed error tolerance. Extensive numerical experiments demonstrate that our approach provides significant acceleration and maintains high numerical stability compared to existing state-of-the-art implementations. These results establish the proposed method as a highly efficient tool for large-scale generative modeling.

## Samyama: A Unified Graph-Vector Database with In-Database Optimization, Agentic Enrichment, and Hardware Acceleration

DevFeed: [Samyama: A Unified Graph-Vector Database with In-Database Optimization, Agentic Enrichment, and Hardware Acceleration](<https://devfeed.tech/articles/samyama-a-unified-graph-vector-database-with-in-database-optimization-agentic-enrichment-and-hardware-acceleration-135433.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.08036>)

Author: Madhulatha Mandarapu, Sandeep Kunkunuru

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [graph-database](<https://devfeed.tech/topics/graph-database.md>), [Vectors & LLMs](<https://devfeed.tech/topics/vectors-llms.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [RDF (Resource Description Framework)](<https://devfeed.tech/topics/rdf.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [RAPIDS](<https://devfeed.tech/topics/rapids.md>)

Tags: [22](<https://devfeed.tech/tags/22.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [cypher](<https://devfeed.tech/tags/cypher.md>), [database](<https://devfeed.tech/tags/database.md>), [database-optimization](<https://devfeed.tech/tags/database-optimization.md>), [databases](<https://devfeed.tech/tags/databases.md>), [edge](<https://devfeed.tech/tags/edge.md>), [enrichment](<https://devfeed.tech/tags/enrichment.md>), [graph](<https://devfeed.tech/tags/graph.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [hardware-acceleration](<https://devfeed.tech/tags/hardware-acceleration.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [vector](<https://devfeed.tech/tags/vector.md>), [vector-database](<https://devfeed.tech/tags/vector-database.md>)

### AI overview

Samyama combines graph and vector storage, analytics, RDF/SPARQL, 22 query-callable metaheuristic solvers, and LLM-based graph enrichment in a Rust engine. At v1.8.0, tests on rented Linux hosts report 123K-152K nodes/s ingestion and PageRank iterations of 6.7 ms at 10,000 nodes and 236 ms at one million on a 16-vCPU instance. On an NVIDIA A16, CUDA matches CPU performance at 10,000 nodes and reaches 1.62x and 2.70x at 100,000 and one million, respectively, with 1.1-1.35 seconds of first-call context initialization. The abstract withdraws earlier GPU, Cypher-throughput, index-scaling, and vector-search figures as unreproducible. It reports 99.9% of evaluated openCypher TCK scenarios passing; in a same-host comparison with Neo4j 5 and FalkorDB, all 14 complex reads were within 5x of the best competitor and 7 were faster.

### Source excerpt

arXiv:2603.08036v4 Announce Type: replace Abstract: Modern data architectures fragment across graph databases, vector stores, analytics engines and optimization solvers, forcing ETL between them. We present Samyama, a graph-vector database in Rust that unifies these workloads in one engine: a RocksDB-backed store with MVCC, a vectorized executor, a cost-based planner, a CSR analytics engine, RDF and SPARQL, 22 metaheuristic solvers callable from the query language, HNSW vector indexing, and agentic enrichment that expands a graph via LLMs. It has been run to billion-edge scale: 74.3M nodes and 1.07B edges from four biomedical sources on one machine for $2.50 of spot compute. This version re-measures the system at release v1.8.0 on rented Linux hosts a reader can boot, replacing earlier figures taken on a Mac Mini that nobody could re-run. On a 16-vCPU cloud instance, ingestion reaches 123K-152K nodes/s and PageRank costs 6.7 ms per iteration at ten thousand nodes and 236 ms at a million. The previously reported 8.2x GPU speedup is withdrawn; in its place we measure the CUDA path, which no earlier version measured at all, on a rented NVIDIA A16: parity with the CPU at ten thousand nodes, 1.62x at a hundred thousand and 2.70x at a million, with 1.1-1.35 s of context initialisation on the first call. The earlier Cypher-throughput figure, its claim of near-constant index-driven scaling, and a vector-search figure are withdrawn as unreproducible. openCypher conformance, once estimated at ~90%, is now measured at 99.9% of evaluated TCK scenarios. Against Neo4j 5 and FalkorDB on one host over an identical SF10 extract, 14 of 14 SNB complex reads land within 5x of the best competitor and 7 of 14 are faster.

[Next page](<https://devfeed.tech/tags/acceleration.md?cursor=WyIyMDI2LTExLTE2VDA1OjAwOjAwKzAwOjAwIiwgIjFmOTRmMjlmLWM4ZTEtNDc4Yy1iN2VhLTQwYmJiNWNkYjVkMCJd>)