# AI Factory

Specialized computing infrastructure optimized for the AI lifecycle, including data ingestion, training, fine-tuning, and high-volume inference.

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

## Inference pushes AI to smaller sites, making the data layer the deciding factor

DevFeed: [Inference pushes AI to smaller sites, making the data layer the deciding factor](<https://devfeed.tech/articles/inference-pushes-ai-to-smaller-sites-making-the-data-layer-the-deciding-factor-146669.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/20/inference-monolithic-data-center-ai-data-layer-sc26/>)

Author: Sloane Kali Faye

Published: 2026-11-20T20:09:20Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [data centers](<https://devfeed.tech/topics/data-centers.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [accelerated-computing](<https://devfeed.tech/tags/accelerated-computing.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-data-layer](<https://devfeed.tech/tags/ai-data-layer.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [alex-bouzari](<https://devfeed.tech/tags/alex-bouzari.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [data-centers](<https://devfeed.tech/tags/data-centers.md>), [data-layer](<https://devfeed.tech/tags/data-layer.md>), [ddn](<https://devfeed.tech/tags/ddn.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-inference](<https://devfeed.tech/tags/distributed-inference.md>), [edge-data-centers](<https://devfeed.tech/tags/edge-data-centers.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprise-ai](<https://devfeed.tech/tags/enterprise-ai.md>), [enterprise-ai-hyperpod](<https://devfeed.tech/tags/enterprise-ai-hyperpod.md>), [exascaler](<https://devfeed.tech/tags/exascaler.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [heterogeneous-computing](<https://devfeed.tech/tags/heterogeneous-computing.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [high-performance-storage](<https://devfeed.tech/tags/high-performance-storage.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [hpc-infrastructure](<https://devfeed.tech/tags/hpc-infrastructure.md>), [inference](<https://devfeed.tech/tags/inference.md>), [john-furrier](<https://devfeed.tech/tags/john-furrier.md>), [news](<https://devfeed.tech/tags/news.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [parallel-file-systems](<https://devfeed.tech/tags/parallel-file-systems.md>), [rebecca-knight](<https://devfeed.tech/tags/rebecca-knight.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [sc26eventpage](<https://devfeed.tech/tags/sc26eventpage.md>), [scientific-computing](<https://devfeed.tech/tags/scientific-computing.md>), [scientific-simulation](<https://devfeed.tech/tags/scientific-simulation.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [thecube](<https://devfeed.tech/tags/thecube.md>)

### AI overview

The excerpt argues that inference is shifting AI infrastructure toward smaller sites near users and data, increasing the importance of a flexible data layer as GPU, memory and SSD costs and shortages constrain budgets. DataDirect Networks CEO Alex Bouzari says software that spans different hardware can help maximize utilization across large AI facilities and smaller 1-5 MW sites. DDN's EXAScaler move to hardware independence reflects that approach; its Enterprise AI HyperPOD combines CPUs, GPUs, networking and storage in a rack for deployment at enterprise sites with available power. The excerpt presents on-premises pods as a way to keep tighter control of company data, and frames data infrastructure as increasingly important alongside accelerated compute. These are Bouzari's views and DDN examples, not an independent comparison of providers.

### Source excerpt

Inference is moving to the center of enterprise AI spending, and the AI data layer is emerging as the piece that decides whether those investments pay off. As GPU costs climb and component shortages persist, the architecture behind AI is shifting from a handful of massive facilities toward a distributed web of smaller sites. That [...] The post Inference pushes AI to smaller sites, making the data layer the deciding factor appeared first on SiliconANGLE.

## D-Matrix bets stacked DRAM can solve inference's memory bandwidth problem

DevFeed: [D-Matrix bets stacked DRAM can solve inference's memory bandwidth problem](<https://devfeed.tech/articles/d-matrix-bets-stacked-dram-can-solve-inference-s-memory-bandwidth-problem-144257.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/19/raptor-chip-stacks-3d-dram-tackle-inference-memory-limits-sc26/>)

Author: Kelly Knight

Published: 2026-11-19T19:21:48Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [d-matrix](<https://devfeed.tech/topics/d-matrix.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [3d-dram](<https://devfeed.tech/tags/3d-dram.md>), [accelerated-computing](<https://devfeed.tech/tags/accelerated-computing.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [can](<https://devfeed.tech/tags/can.md>), [chip](<https://devfeed.tech/tags/chip.md>), [compute](<https://devfeed.tech/tags/compute.md>), [corsair](<https://devfeed.tech/tags/corsair.md>), [cube-event-coverage](<https://devfeed.tech/tags/cube-event-coverage.md>), [d-matrix](<https://devfeed.tech/tags/d-matrix.md>), [designs](<https://devfeed.tech/tags/designs.md>), [dram](<https://devfeed.tech/tags/dram.md>), [emerging-tech](<https://devfeed.tech/tags/emerging-tech.md>), [first](<https://devfeed.tech/tags/first.md>), [game](<https://devfeed.tech/tags/game.md>), [generation](<https://devfeed.tech/tags/generation.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [high-bandwidth-memory](<https://devfeed.tech/tags/high-bandwidth-memory.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [high-speed-networking](<https://devfeed.tech/tags/high-speed-networking.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [hpc-infrastructure](<https://devfeed.tech/tags/hpc-infrastructure.md>), [inference](<https://devfeed.tech/tags/inference.md>), [john-furrier](<https://devfeed.tech/tags/john-furrier.md>), [memory](<https://devfeed.tech/tags/memory.md>), [memory-bandwidth](<https://devfeed.tech/tags/memory-bandwidth.md>), [nasdaq-nvda](<https://devfeed.tech/tags/nasdaq-nvda.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvlink-fusion](<https://devfeed.tech/tags/nvlink-fusion.md>), [raptor](<https://devfeed.tech/tags/raptor.md>), [satyam-srivastava](<https://devfeed.tech/tags/satyam-srivastava.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [sc26eventpage](<https://devfeed.tech/tags/sc26eventpage.md>), [scientific-simulation](<https://devfeed.tech/tags/scientific-simulation.md>), [solve](<https://devfeed.tech/tags/solve.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [sustainable-computing](<https://devfeed.tech/tags/sustainable-computing.md>), [thecube](<https://devfeed.tech/tags/thecube.md>)

### AI overview

D-Matrix says Raptor, its planned second-generation chip, will stack 3D DRAM beneath compute logic to address inference's memory-bandwidth demands. The Pavehawk test chip is validating the approach: memory sits on a separate layer, close to logic, without crossing an interposer. The company places this memory tier between SRAM and HBM in capacity and claims 10 times HBM's bandwidth and energy efficiency. Srivastava also says pairing Corsair-based servers with Hopper GPUs can raise utilization from about 20% to as much as 80%. D-Matrix's adoption of Nvidia NVLink Fusion is intended to connect its accelerators to Nvidia AI factories, supporting systems that coordinate GPUs, CPUs and accelerators according to workload. These performance figures and utilization gains are company-reported claims.

### Source excerpt

As artificial intelligence workloads shift to inference, memory bandwidth decides whether hardware generates tokens quickly or sits idle. D-Matrix Corp. is answering with Raptor, a second-generation chip stacking 3D DRAM beneath its compute logic. AI infrastructure has become a systems game, and winners combine memory, interconnect and software into cohesive designs, according to Satyam Srivastava [...] The post D-Matrix bets stacked DRAM can solve inference's memory bandwidth problem appeared first on SiliconANGLE.

## Today's data centers weren't built for AI racks, so liquid-cooling sidecars get bolted on

DevFeed: [Today's data centers weren't built for AI racks, so liquid-cooling sidecars get bolted on](<https://devfeed.tech/articles/today-s-data-centers-weren-t-built-for-ai-racks-so-liquid-cooling-sidecars-get-bolted-on-143692.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/19/deployment-services-drive-supermicro-s-race-time-online-sc26/>)

Author: Jonathan Anthony

Published: 2026-11-19T14:02:06Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [data centers](<https://devfeed.tech/topics/data-centers.md>)

Tags: [accelerated-computing](<https://devfeed.tech/tags/accelerated-computing.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-factories](<https://devfeed.tech/tags/ai-factories.md>), [ai-factory](<https://devfeed.tech/tags/ai-factory.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-supercomputers](<https://devfeed.tech/tags/ai-supercomputers.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [cooling](<https://devfeed.tech/tags/cooling.md>), [cube-event-coverage](<https://devfeed.tech/tags/cube-event-coverage.md>), [data-center-retrofits](<https://devfeed.tech/tags/data-center-retrofits.md>), [data-centers](<https://devfeed.tech/tags/data-centers.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [deployment-services](<https://devfeed.tech/tags/deployment-services.md>), [energy-efficiency](<https://devfeed.tech/tags/energy-efficiency.md>), [gemma-allen](<https://devfeed.tech/tags/gemma-allen.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [hpc-infrastructure](<https://devfeed.tech/tags/hpc-infrastructure.md>), [hpc-unites](<https://devfeed.tech/tags/hpc-unites.md>), [infra](<https://devfeed.tech/tags/infra.md>), [john-furrier](<https://devfeed.tech/tags/john-furrier.md>), [liquid-cooling](<https://devfeed.tech/tags/liquid-cooling.md>), [matthew-thauberger](<https://devfeed.tech/tags/matthew-thauberger.md>), [nasdaq-smci](<https://devfeed.tech/tags/nasdaq-smci.md>), [neoclouds](<https://devfeed.tech/tags/neoclouds.md>), [news](<https://devfeed.tech/tags/news.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [sc26-2026](<https://devfeed.tech/tags/sc26-2026.md>), [sc26eventpage](<https://devfeed.tech/tags/sc26eventpage.md>), [sidecars](<https://devfeed.tech/tags/sidecars.md>), [super-micro-computer](<https://devfeed.tech/tags/super-micro-computer.md>), [supermicro](<https://devfeed.tech/tags/supermicro.md>), [sustainable-computing](<https://devfeed.tech/tags/sustainable-computing.md>), [thecube](<https://devfeed.tech/tags/thecube.md>)

### AI overview

With GPU supply constrained, buyers are judging AI infrastructure providers by how quickly systems reach production. Supermicro says its teams handle allocation, on-site deployment, workload testing and handoff once systems are running. For older facilities not designed for the power density and liquid cooling of modern AI racks, it offers retrofit options including sidecars and in-row coolant distribution units; pre-assembled AI racks are another way it aims to shorten installation. These are the company's stated offerings and rationale, as reported in an interview.

### Source excerpt

Deployment services are becoming the real battleground in AI infrastructure as buyers discover that shipping hardware is only half the job. With GPU supply tight and refresh cycles accelerating, neoclouds and enterprises are judging suppliers on how quickly an AI factory actually goes live. Server makers that once competed on time to market are now [...] The post Today's data centers weren't built for AI racks, so liquid-cooling sidecars get bolted on appeared first on SiliconANGLE.

## CoreWeave Tops NVIDIA Blackwell Inference Throughput, Cuts TCO Up to 65%

DevFeed: [CoreWeave Tops NVIDIA Blackwell Inference Throughput, Cuts TCO Up to 65%](<https://devfeed.tech/articles/coreweave-tops-nvidia-blackwell-inference-throughput-cuts-tco-up-to-65-142075.md>)

Original publisher: [Read original article](<https://wf.coreweave.com/blog/coreweave-tops-nvidia-blackwell-inference-throughput-cuts-tco-up-to-65>)

Author: CoreWeave

Published: 2026-11-18T15:38:38Z

Content type: news

Language: en

Sources: [Coreweave Blog Posts](<https://devfeed.tech/sources/coreweave-blog-posts.md>)

Topics: [NVIDIA Blackwell](<https://devfeed.tech/topics/nvidia-blackwell.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [GB200](<https://devfeed.tech/topics/gb200.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>)

Tags: [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [api](<https://devfeed.tech/tags/api.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [bare-metal](<https://devfeed.tech/tags/bare-metal.md>), [blackwell](<https://devfeed.tech/tags/blackwell.md>), [blog](<https://devfeed.tech/tags/blog.md>), [cks](<https://devfeed.tech/tags/cks.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-infrastructure](<https://devfeed.tech/tags/cloud-infrastructure.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [coreweave](<https://devfeed.tech/tags/coreweave.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cost-per-token](<https://devfeed.tech/tags/cost-per-token.md>), [costs](<https://devfeed.tech/tags/costs.md>), [inference](<https://devfeed.tech/tags/inference.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvidia-blackwell](<https://devfeed.tech/tags/nvidia-blackwell.md>), [scale](<https://devfeed.tech/tags/scale.md>), [tco](<https://devfeed.tech/tags/tco.md>), [throughput](<https://devfeed.tech/tags/throughput.md>), [tokens](<https://devfeed.tech/tags/tokens.md>)

### AI overview

Signal65's modeled 5,000-GPU, three-year comparison reports CoreWeave had the lowest TCO in six of seven configurations and the highest per-GPU inference throughput among the four clouds evaluated. For GB200 NVL72, its modeled cost was 22% to 65% below the hyperscalers; the 22% comparison uses an estimated Hyperscaler B price. On DeepSeek-R1, CoreWeave delivered 4,867 tokens per second per GPU and $0.75 per million tokens, versus $1.23-$2.22 elsewhere, yielding up to 2.9 times as many tokens per dollar. The account says the model omits some hyperscaler storage fees and archive deletion penalties. For hosted inference, CoreWeave was cheaper on 11 of 17 comparable models and tied on four; on gpt-oss-120b, its input and output token prices were respectively 67%-80% and 53%-72% lower. The results are reported in a CoreWeave-authored summary of the study.

### Source excerpt

What does NVIDIA Blackwell cost at scale? Signal65 found CoreWeave delivered the lowest three-year TCO in six of seven configurations and up to 2.9x more tokens per dollar.

## Stalled data center buildouts send inference chips hunting for space in telco and colo sites

DevFeed: [Stalled data center buildouts send inference chips hunting for space in telco and colo sites](<https://devfeed.tech/articles/stalled-data-center-buildouts-send-inference-chips-hunting-for-space-in-telco-and-colo-sites-140558.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/18/ai-data-centers-delays-push-inference-chips-existing-sites-sc26/>)

Author: Jonathan Anthony

Published: 2026-11-18T12:13:22Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [data centers](<https://devfeed.tech/topics/data-centers.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [accelerated-computing](<https://devfeed.tech/tags/accelerated-computing.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [anton-mcgonnell](<https://devfeed.tech/tags/anton-mcgonnell.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [capital](<https://devfeed.tech/tags/capital.md>), [chipmakers](<https://devfeed.tech/tags/chipmakers.md>), [chips](<https://devfeed.tech/tags/chips.md>), [colocation](<https://devfeed.tech/tags/colocation.md>), [construction](<https://devfeed.tech/tags/construction.md>), [cube-event-coverage](<https://devfeed.tech/tags/cube-event-coverage.md>), [d-series-gpu](<https://devfeed.tech/tags/d-series-gpu.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [data-centers](<https://devfeed.tech/tags/data-centers.md>), [demand](<https://devfeed.tech/tags/demand.md>), [energy-efficiency](<https://devfeed.tech/tags/energy-efficiency.md>), [existing](<https://devfeed.tech/tags/existing.md>), [facilities](<https://devfeed.tech/tags/facilities.md>), [gemma-allen](<https://devfeed.tech/tags/gemma-allen.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [hpc-infrastructure](<https://devfeed.tech/tags/hpc-infrastructure.md>), [hpc-unites](<https://devfeed.tech/tags/hpc-unites.md>), [inference](<https://devfeed.tech/tags/inference.md>), [inference-chips](<https://devfeed.tech/tags/inference-chips.md>), [infra](<https://devfeed.tech/tags/infra.md>), [intel](<https://devfeed.tech/tags/intel.md>), [john-furrier](<https://devfeed.tech/tags/john-furrier.md>), [liquid](<https://devfeed.tech/tags/liquid.md>), [news](<https://devfeed.tech/tags/news.md>), [power](<https://devfeed.tech/tags/power.md>), [research-computing](<https://devfeed.tech/tags/research-computing.md>), [room](<https://devfeed.tech/tags/room.md>), [sambanova](<https://devfeed.tech/tags/sambanova.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [sc26-2026](<https://devfeed.tech/tags/sc26-2026.md>), [sc26eventpage](<https://devfeed.tech/tags/sc26eventpage.md>), [sn50](<https://devfeed.tech/tags/sn50.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [sustainable-computing](<https://devfeed.tech/tags/sustainable-computing.md>), [telco](<https://devfeed.tech/tags/telco.md>), [thecube](<https://devfeed.tech/tags/thecube.md>)

### AI overview

Inference demand is growing faster than new data center capacity because power access, financing, permitting and equipment lead times slow construction. Older telco and colocation sites may have space and power, but many cannot support dense liquid-cooled GPU racks without costly retrofits. SambaNova says its SN50 decode racks are air-cooled and light enough for existing facilities; its SC26 demonstration paired them with Intel D-Series GPUs for prefill in a heterogeneous setup. The article contrasts these racks with newer GPU racks that may require reinforced floors and 120-250 kW per rack. It reports 18-24-month lead times for power and cooling equipment, and attributes the SN50's fit to its focus on decode, where data movement is a bottleneck.

### Source excerpt

Demand for inference is outrunning the construction of new AI data centers, as power, capital and permitting delays push fresh capacity further out. That gap is sending chipmakers to hunt for room in existing facilities. Many older telco and colocation sites have the space and power. But most can't take today's dense, liquid-cooled GPU racks [...] The post Stalled data center buildouts send inference chips hunting for space in telco and colo sites appeared first on SiliconANGLE.

## What Production AI Demands from the Data Layer

DevFeed: [What Production AI Demands from the Data Layer](<https://devfeed.tech/articles/what-production-ai-demands-from-the-data-layer-139722.md>)

Original publisher: [Read original article](<https://blog.everpuredata.com/solutions/what-production-ai-demands-from-the-data-layer/>)

Author: Sabrinath Rao

Published: 2026-11-17T13:50:00Z

Content type: opinion

Language: en

Sources: [Everpure](<https://devfeed.tech/sources/everpure.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [scalable-architecture](<https://devfeed.tech/topics/scalable-architecture.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [Cache](<https://devfeed.tech/topics/cache.md>), [IO](<https://devfeed.tech/topics/io.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-and-machine-learning](<https://devfeed.tech/tags/ai-and-machine-learning.md>), [ai-data](<https://devfeed.tech/tags/ai-data.md>), [ai-data-layer](<https://devfeed.tech/tags/ai-data-layer.md>), [ai-training](<https://devfeed.tech/tags/ai-training.md>), [ai-training-and-inference](<https://devfeed.tech/tags/ai-training-and-inference.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [attention](<https://devfeed.tech/tags/attention.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [bottleneck](<https://devfeed.tech/tags/bottleneck.md>), [cache](<https://devfeed.tech/tags/cache.md>), [data](<https://devfeed.tech/tags/data.md>), [data-layer](<https://devfeed.tech/tags/data-layer.md>), [flashblade-exa](<https://devfeed.tech/tags/flashblade-exa.md>), [flashblade-object-storage](<https://devfeed.tech/tags/flashblade-object-storage.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [inference](<https://devfeed.tech/tags/inference.md>), [production-ai](<https://devfeed.tech/tags/production-ai.md>), [solutions](<https://devfeed.tech/tags/solutions.md>)

### AI overview

The post argues that production AI benefits when storage resources scale independently and are measured by their effect on useful compute. FlashBlade//EXA separates metadata operations from payload traffic, allowing each path to scale on demand; STN reports growth from about 35 GB/s to as much as 100 GB/s. In submitted MLPerf Storage v3.0 results, it ranked first on the reported checkpointing and KV cache metrics, with checkpoint write bandwidth rising from 327.65 GiB/s on 10 data nodes to 877.52 GiB/s on 30; the author cautions these workloads are proxies and recommends testing real workload mixes. Purity//FB 4.8.4 adds S3 over RDMA through standard S3 calls, though the newer capability has early-release constraints. Pure KVA retains attention states and integrates with vLLM; company benchmarks report faster inference and improved time to first token. For multi-tenancy, file systems are assigned to separate data node groups to isolate drives, ports, and failures. The suggested evaluations include checkpoint stall time, goodput, concurrent time to first token, data copies, node-failure impact, and watts per sustained GB/s.

### Source excerpt

What Production AI Demands from the Data Layer by Everpure Blog Explore how a scalable AI data layer can accelerate checkpointing and inference while helping keep GPUs productive across AI workloads. The post What Production AI Demands from the Data Layer appeared first on Everpure Blog.

## Hedra Grows Revenue 15x and Opens Its Model Factory to Build the Brains of Physical AI

DevFeed: [Hedra Grows Revenue 15x and Opens Its Model Factory to Build the Brains of Physical AI](<https://devfeed.tech/articles/hedra-grows-revenue-15x-and-opens-its-model-factory-to-build-the-brains-of-physical-ai-127608.md>)

Original publisher: [Read original article](<https://hackernoon.com/hedra-grows-revenue-15x-and-opens-its-model-factory-to-build-the-brains-of-physical-ai?source=rss>)

Author: Ishan Pandey

Published: 2026-11-10T18:08:35Z

Content type: news

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>)

Tags: [a16z](<https://devfeed.tech/tags/a16z.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-and-ml](<https://devfeed.tech/tags/ai-and-ml.md>), [biology](<https://devfeed.tech/tags/biology.md>), [business](<https://devfeed.tech/tags/business.md>), [defense](<https://devfeed.tech/tags/defense.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [hedra](<https://devfeed.tech/tags/hedra.md>), [inference](<https://devfeed.tech/tags/inference.md>), [launches](<https://devfeed.tech/tags/launches.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [physical-ai](<https://devfeed.tech/tags/physical-ai.md>), [platform](<https://devfeed.tech/tags/platform.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [revenue](<https://devfeed.tech/tags/revenue.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [startups](<https://devfeed.tech/tags/startups.md>), [technology](<https://devfeed.tech/tags/technology.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

Hedra says its Model Factory combines custom post-training, inference deployment, and ongoing model improvement for robotics, manufacturing, biology, and defense. Customers can bring simulation data or an existing model; Hedra researchers and AI agents help build training environments, tune models, and deploy them on premises or at a nearby partner data center. Production models are monitored and retrained, with releases gated by each customer's evaluations. Hedra reports a concrete speed result with Krea 3: reducing 2K image edit time from about 22 seconds to 3.6 seconds on the same four AMD Instinct MI355X GPUs through distillation and inference optimization. The article also reports a robot gripper simulation in which task success recovered from near zero to near 100% without retraining. Hedra says revenue grew 15x over the past year; the platform is available at hedra.com.

### Source excerpt

Hedra grows revenue 15x and launches Model Factory, a self-improving post-training and inference platform for robotics, manufacturing, biology and defense.

## CNCF Welcomes Vultr as a Platinum Member to Advance Open and Sovereign Cloud Infrastructure for AI

DevFeed: [CNCF Welcomes Vultr as a Platinum Member to Advance Open and Sovereign Cloud Infrastructure for AI](<https://devfeed.tech/articles/cncf-welcomes-vultr-as-a-platinum-member-to-advance-open-and-sovereign-cloud-infrastructure-for-ai-127627.md>)

Original publisher: [Read original article](<https://www.cncf.io/announcements/2026/11/10/cncf-welcomes-vultr-as-a-platinum-member-to-advance-open-and-sovereign-cloud-infrastructure-for-ai/>)

Author: kthornhill

Published: 2026-11-10T16:00:00Z

Content type: news

Language: en

Sources: [Cloud Native Computing Foundation](<https://devfeed.tech/sources/cloud-native-computing-foundation.md>)

Topics: [Cloud Native Ecosystem](<https://devfeed.tech/topics/cloud-native-ecosystem.md>), [Sovereign AI](<https://devfeed.tech/topics/sovereign-ai.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-factory](<https://devfeed.tech/tags/ai-factory.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-native](<https://devfeed.tech/tags/cloud-native.md>), [cloud-native-ai-infrastructure](<https://devfeed.tech/tags/cloud-native-ai-infrastructure.md>), [cloud-native-computing-foundation](<https://devfeed.tech/tags/cloud-native-computing-foundation.md>), [cloudnativecon](<https://devfeed.tech/tags/cloudnativecon.md>), [cncf](<https://devfeed.tech/tags/cncf.md>), [community](<https://devfeed.tech/tags/community.md>), [contributed](<https://devfeed.tech/tags/contributed.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [interoperability](<https://devfeed.tech/tags/interoperability.md>), [kubecon](<https://devfeed.tech/tags/kubecon.md>), [kubecon-plus-cloudnativecon-north-america](<https://devfeed.tech/tags/kubecon-plus-cloudnativecon-north-america.md>), [north-america](<https://devfeed.tech/tags/north-america.md>), [open](<https://devfeed.tech/tags/open.md>), [sovereign-cloud](<https://devfeed.tech/tags/sovereign-cloud.md>), [vultr](<https://devfeed.tech/tags/vultr.md>)

### AI overview

Vultr has upgraded to Platinum membership in the Cloud Native Computing Foundation (CNCF), building on its membership since 2021 and its work with Kubernetes and Kubeflow maintainers. The company says its engineers contribute feedback and improvements upstream, its Vultr Kubernetes Engine is CNCF Certified, and it has supplied CPU and GPU capacity for CNCF project testing. Vultr also launched the Vultr Open Source Alliance and supports open-source frameworks including PyTorch for model training and Kubernetes for inference. The announcement frames these efforts as supporting open, interoperable AI infrastructure that lets developers choose their tools and infrastructure; it provides no details on the amount or results of the testing capacity contribution.

### Source excerpt

Upgrade extends Vultr's 5-year history with CNCF and expands its voice in shaping open interoperability for AI factory infrastructure at global scale Key Highlights KUBECON + CLOUDNATIVECON NORTH AMERICA, SALT LAKE CITY - November 10, 2026-...

## Monetizing compliance: New AI revenue streams for regional CSPs

DevFeed: [Monetizing compliance: New AI revenue streams for regional CSPs](<https://devfeed.tech/articles/monetizing-compliance-new-ai-revenue-streams-for-regional-csps-126758.md>)

Original publisher: [Read original article](<https://www.cio.com/article/4246916/monetizing-compliance-new-ai-revenue-streams-for-regional-csps.html>)

Author: HPE and AMD

Published: 2026-11-10T15:29:10Z

Content type: article

Language: en

Sources: [CIO](<https://devfeed.tech/sources/cio.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Advantages and Disadvantages of Cloud Service Models](<https://devfeed.tech/topics/advantages-and-disadvantages-of-cloud-service-models.md>), [control-plane](<https://devfeed.tech/topics/control-plane.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [ai-revenue](<https://devfeed.tech/tags/ai-revenue.md>), [ai-services](<https://devfeed.tech/tags/ai-services.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [applications](<https://devfeed.tech/tags/applications.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [banking](<https://devfeed.tech/tags/banking.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [csp](<https://devfeed.tech/tags/csp.md>), [customer-data](<https://devfeed.tech/tags/customer-data.md>), [customization](<https://devfeed.tech/tags/customization.md>), [data](<https://devfeed.tech/tags/data.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprise-cloud](<https://devfeed.tech/tags/enterprise-cloud.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [revenue](<https://devfeed.tech/tags/revenue.md>), [storytype-brandpost](<https://devfeed.tech/tags/storytype-brandpost.md>), [streams](<https://devfeed.tech/tags/streams.md>)

### AI overview

Regional CSPs can offer sovereign AI services to organizations whose sensitive data or regulatory obligations limit use of public AI. The proposed revenue streams include localized LLM inference, model customization on proprietary data, and industry-specific applications; the article argues that compliance and isolation can support premium pricing. It identifies computational efficiency and tenant management as key challenges, and describes HPE systems with 5th Gen AMD EPYC processors and AMD Instinct GPUs as a purpose-built infrastructure option. The text reports benchmark gains of up to 1.73x for end-to-end AI and 1.93x for machine learning throughput versus competing x86 options. It cites Finland's LUMI and France's Adastra as sovereign deployments, and names HPE Morpheus for orchestration and tenant isolation, HPE OpsRamp for infrastructure monitoring, resource tracking for usage-based billing, and air-gapped management for disconnected environments.

### Source excerpt

The rush to adopt artificial intelligence has hit a regulatory and geopolitical wall. For governments, financial institutions, healthcare networks, and critical infrastructure operators, standard public AI models present risks. Sending sensitive data across borders or exposing proprietary data to multitenant public environments can violate compliance mandates. This tension creates an extraordinary market opportunity for cloud service providers (CSPs). By launching sovereign AI services--environments where data, models, and execution remain strictly within national, regional, or specific legal jurisdictions--local and regional CSPs can capture a highly lucrative, rapidly expanding segment of enterprise cloud spend. Monetizing this trend requires moving beyond standard infrastructure provisioning. It demands high-performance, legally isolated environments engineered to handle intense AI workloads without sacrificing control. The business case for sovereign AI Traditional public clouds thrive on global scale, but sovereign AI relies on local trust. For CSPs, building a sovereign AI offering allows them to capture revenue from organizations that previously could not use AI due to legal restrictions. By operating a sovereign platform, CSPs can package, resell, and manage multiple value-added services: Enterprise inferencing: Localized deployment of large language models (LLMs) for specific data querying. Model customization: Fine-tuning base open-source models using highly classified or proprietary customer data. Industry-specific AI applications: Compliant platforms customized for regional healthcare codes or localized banking laws. CSPs can position themselves as trusted partners rather than just infrastructure vendors, commanding premium margins for compliance and isolation. Overcoming architectural hurdles with purpose-built AI infrastructure To successfully launch a sovereign AI business, a CSP must overcome two primary architectural challenges: raw computational effi

## Google TPU 8tx Pairs AMD EPYC Turin Hosts with Confidential Computing while TPU 8i Max Doubles Memory Bandwidth

DevFeed: [Google TPU 8tx Pairs AMD EPYC Turin Hosts with Confidential Computing while TPU 8i Max Doubles Memory Bandwidth](<https://devfeed.tech/articles/google-tpu-8tx-pairs-amd-epyc-turin-hosts-with-confidential-computing-while-tpu-8i-max-doubles-memory-bandwidth-127157.md>)

Original publisher: [Read original article](<https://www.servethehome.com/google-tpu-8tx-8i-max-amd-epyc-turin/>)

Author: Vic A

Published: 2026-11-09T21:37:58Z

Content type: article

Language: en

Sources: [ServeTheHome](<https://devfeed.tech/sources/servethehome.md>)

Topics: [Confidential Computing](<https://devfeed.tech/topics/confidential-computing.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [Security & Privacy](<https://devfeed.tech/topics/security-privacy.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-workloads](<https://devfeed.tech/tags/agentic-workloads.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [amd](<https://devfeed.tech/tags/amd.md>), [amd-epyc](<https://devfeed.tech/tags/amd-epyc.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [confidential-computing](<https://devfeed.tech/tags/confidential-computing.md>), [epyc](<https://devfeed.tech/tags/epyc.md>), [google](<https://devfeed.tech/tags/google.md>), [google-tpu](<https://devfeed.tech/tags/google-tpu.md>), [google-tpu-8tx](<https://devfeed.tech/tags/google-tpu-8tx.md>), [inference](<https://devfeed.tech/tags/inference.md>), [memory](<https://devfeed.tech/tags/memory.md>), [memory-bandwidth](<https://devfeed.tech/tags/memory-bandwidth.md>), [server](<https://devfeed.tech/tags/server.md>), [tpu](<https://devfeed.tech/tags/tpu.md>), [tpu-8i](<https://devfeed.tech/tags/tpu-8i.md>), [tpu-8i-max](<https://devfeed.tech/tags/tpu-8i-max.md>), [tpu-8tx](<https://devfeed.tech/tags/tpu-8tx.md>), [turin](<https://devfeed.tech/tags/turin.md>)

### AI overview

Google's TPU 8tx and TPU 8i Max target different needs using existing TPU silicon. The 8tx pairs TPU 8t silicon with 5th Gen AMD EPYC Turin hosts and confidential computing, which protects data while it is being processed. Its dual-socket hosts provide 128 PCIe Gen5 lanes for four TPUs and high-speed NICs; Google says its Boardfly network can deliver up to 2x lower latency than TPU 8t with near-zero performance penalty. For inference and serving where data must stay within the confidential computing domain, Google recommends 8tx. The 8i Max combines two TPU 8i chips in one package, doubling compute, SRAM and HBM bandwidth versus one 8i; it provides 384 GB HBM, 17.2 TB/s bandwidth and 768 MB SRAM. Both use Boardfly, while the article distinguishes 8i Max as an inference-focused option for memory-intensive workloads.

### Source excerpt

The new Google TPU 8tx pairs AMD EPYC Turin hosts with confidential computing while TPU 8i Max doubles memory bandwidth for inference The post Google TPU 8tx Pairs AMD EPYC Turin Hosts with Confidential Computing while TPU 8i Max Doubles Memory Bandwidth appeared first on ServeTheHome.

## Google Rejiggers TPU Accelerators For Broader AI Inference Work

DevFeed: [Google Rejiggers TPU Accelerators For Broader AI Inference Work](<https://devfeed.tech/articles/google-rejiggers-tpu-accelerators-for-broader-ai-inference-work-124816.md>)

Original publisher: [Read original article](<https://www.nextplatform.com/compute/2026/11/09/google-rejiggers-tpu-accelerators-for-broader-ai-inference-work/5308534>)

Author: Timothy Prickett Morgan

Published: 2026-11-09T18:06:49Z

Content type: news

Language: en

Sources: [The Next Platform: In-depth coverage of high end computing](<https://devfeed.tech/sources/the-next-platform-in-depth-coverage-of-high-end-computing.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [Google Cloud Platform (GCP)](<https://devfeed.tech/topics/google-cloud.md>), [Google](<https://devfeed.tech/topics/google.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [broadcom](<https://devfeed.tech/tags/broadcom.md>), [compute](<https://devfeed.tech/tags/compute.md>), [google](<https://devfeed.tech/tags/google.md>), [inference](<https://devfeed.tech/tags/inference.md>), [mediatek](<https://devfeed.tech/tags/mediatek.md>), [sunfish](<https://devfeed.tech/tags/sunfish.md>), [tpu](<https://devfeed.tech/tags/tpu.md>), [tpu-8i](<https://devfeed.tech/tags/tpu-8i.md>), [tpu-8t](<https://devfeed.tech/tags/tpu-8t.md>), [zebrafish](<https://devfeed.tech/tags/zebrafish.md>)

### AI overview

Google's new TPU variants target distinct workloads through changes to interconnect, memory, and host processors. The TPU 8tx adapts the training-focused TPU 8t for inference: its boardfly fabric reduces pod size from 9,600 to 1,152 accelerators and is described as having about half the average latency of the 3D torus, while an AMD Epyc replaces Google's Axion CPU to provide hardware root of trust and encryption acceleration. The TPU 8i Max pairs two Zebrafish chips to provide more shared memory for long-context inference, though memory capacity rises only 33% per socket and memory per unit of compute falls. Google's TPU Sync software moves an agent's working memory off the TPU while paused and restores it on resume, freeing accelerators for other tasks; the excerpt says it also supports TPU 5 through TPU 8, but gives no throughput measurements. The modified systems are expected in 2027. Their cloud pricing and the scale and performance of TPU Sync remain uncertain.

### Source excerpt

Sometimes you need a new chip, and sometimes you need a new way to configure the chips you have to b ...

## How Aunalytics built infrastructure for AI-ready data | Spectro Cloud

DevFeed: [How Aunalytics built infrastructure for AI-ready data | Spectro Cloud](<https://devfeed.tech/articles/how-aunalytics-built-infrastructure-for-ai-ready-data-spectro-cloud-123528.md>)

Original publisher: [Read original article](<https://www.spectrocloud.com/blog/aunalytics-infrastructure-ai-ready-data>)

Author: Rob Collins

Published: 2026-11-09T08:00:00Z

Content type: article

Language: en

Sources: [Spectro Cloud Feed](<https://devfeed.tech/sources/spectro-cloud-feed.md>)

Topics: [WTF Is Cloud Native](<https://devfeed.tech/topics/wtf-is-cloud-native.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [hosting](<https://devfeed.tech/topics/hosting.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>)

Tags: [2025](<https://devfeed.tech/tags/2025.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-ready-data](<https://devfeed.tech/tags/ai-ready-data.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [banks](<https://devfeed.tech/tags/banks.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [data](<https://devfeed.tech/tags/data.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [multi-tenant](<https://devfeed.tech/tags/multi-tenant.md>), [palette](<https://devfeed.tech/tags/palette.md>), [team](<https://devfeed.tech/tags/team.md>), [virtual](<https://devfeed.tech/tags/virtual.md>)

### AI overview

Aunalytics says Spectro Cloud's self-hosted Palette helped three engineers scale a multi-tenant cloud from 3 clusters and 15 nodes to 6 clusters, 30 nodes, and more than 40 virtual clusters. MAAS automates bare-metal provisioning; versioned profiles define the OS, Kubernetes, networking, storage, and add-ons. Palette virtual clusters provide separate API servers and RBAC on shared hardware, while dedicated clusters remain available where clients require full separation. Aunalytics reports reducing new-environment setup from about two days to under an hour, rolling out upgrades in two-hour windows without client-facing downtime, and seeing 90%+ fewer configuration-drift incidents. It also runs H100 GPU clusters under the same profiles and tenancy. A phased KubeVirt-based VMO initiative aims to move about 90% of its VM estate over six months, starting with stateless workloads; some VMs will remain on VMware for licensing or vendor-support reasons. These are Aunalytics' reported outcomes and plans.

### Source excerpt

Aunalytics' Rob Collins on how a three-person team runs a multi-tenant analytics cloud for banks and credit unions on Palette, from virtual clusters to VMs.

## MeshRT: Compile-Time Governed Wafer-Scale Runtime for Low-Latency High-Throughput Inference

DevFeed: [MeshRT: Compile-Time Governed Wafer-Scale Runtime for Low-Latency High-Throughput Inference](<https://devfeed.tech/articles/meshrt-compile-time-governed-wafer-scale-runtime-for-low-latency-high-throughput-inference-123672.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.08508>)

Author: Congjie He, Le Xu, Zhan Lu, Yeqi Huang, Haocheng Xiao, Cheng Deng, Lingxiao Ma, Ziming Miao, Fan Yang, Luo Mai

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

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

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [vllm](<https://devfeed.tech/topics/vllm.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [asynchronous](<https://devfeed.tech/tags/asynchronous.md>), [batching](<https://devfeed.tech/tags/batching.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [compile](<https://devfeed.tech/tags/compile.md>), [core](<https://devfeed.tech/tags/core.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-os](<https://devfeed.tech/tags/cs-os.md>), [decentralized](<https://devfeed.tech/tags/decentralized.md>), [decode](<https://devfeed.tech/tags/decode.md>), [events](<https://devfeed.tech/tags/events.md>), [execution](<https://devfeed.tech/tags/execution.md>), [inference](<https://devfeed.tech/tags/inference.md>), [latency](<https://devfeed.tech/tags/latency.md>), [low-latency](<https://devfeed.tech/tags/low-latency.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [scale](<https://devfeed.tech/tags/scale.md>), [throughput](<https://devfeed.tech/tags/throughput.md>)

### AI overview

MeshRT aims to combine low latency with high throughput for LLM inference on wafer-scale accelerators, where runtime-dynamic techniques such as batching and mixture-of-experts can be difficult to support efficiently. Its approach compiles inference events into per-core schedules, enabling asynchronous, decentralized execution with low runtime overhead. On a commodity wafer-scale accelerator, the authors report up to 56x higher throughput than the prior state-of-the-art wafer-scale system. Against measured single-instance SGLang results on an H200, they report 4.5-7.0x lower latency; for decode throughput, they report 1.7-13.0x higher throughput only when those GPU measurements are ideally linearly scaled to an approximately power-matched H200 cluster.

### Source excerpt

arXiv:2611.08508v1 Announce Type: cross Abstract: Wafer-scale accelerators promise ultra-low-latency AI inference, but current system stacks still carry an unsustainably high cost premium. The reason is that many current and emerging inference techniques, such as batching and MoE, introduce runtime dynamism that existing wafer-scale systems cannot support efficiently. As a result, they must choose between two unsatisfactory options: accept suboptimal latency, or preserve low latency by restricting dynamic techniques, thereby limiting the models they can support and reducing aggregate throughput. This tradeoff significantly increases the cost of low-latency serving. We present MeshRT, the first wafer-scale system to achieve both ultra-low latency and high throughput for LLM inference. MeshRT achieves this via a novel system architecture, which we call compile-time governance of runtime dynamism. This architecture features new system abstractions and mechanisms that enable MeshRT to compile dynamic events arising from LLM inference directly into per-core schedules, enabling asynchronous and decentralized execution with low runtime overhead. Implemented on a commodity wafer-scale accelerator, MeshRT preserves ultra-low latency while improving throughput by up to 56$\times$ over the state-of-the-art system for wafer-scale accelerators. Against measured single-instance SGLang results on H200, MeshRT achieves 4.5-7.0$\times$ lower latency. Under ideal linear scaling of those GPU measurements to an approximately power-matched H200 cluster, MeshRT achieves 1.7-13.0$\times$ higher decode system throughput in the low-latency regime.

## Nvidia, Municipal Leaders Frame AI as Critical Urban Infrastructure

DevFeed: [Nvidia, Municipal Leaders Frame AI as Critical Urban Infrastructure](<https://devfeed.tech/articles/nvidia-municipal-leaders-frame-ai-as-critical-urban-infrastructure-119306.md>)

Original publisher: [Read original article](<https://www.eetimes.com/nvidia-municipal-leaders-frame-ai-as-critical-urban-infrastructure/>)

Author: Pablo Valerio

Published: 2026-11-05T16:09:46Z

Content type: news

Language: en

Sources: [EE Times](<https://devfeed.tech/sources/ee-times.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [digital sovereignty](<https://devfeed.tech/topics/digital-sovereignty.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [ai-factories](<https://devfeed.tech/tags/ai-factories.md>), [ai-journey](<https://devfeed.tech/tags/ai-journey.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [air-quality](<https://devfeed.tech/tags/air-quality.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [analog](<https://devfeed.tech/tags/analog.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [application](<https://devfeed.tech/tags/application.md>), [applications](<https://devfeed.tech/tags/applications.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [blog](<https://devfeed.tech/tags/blog.md>), [business](<https://devfeed.tech/tags/business.md>), [city](<https://devfeed.tech/tags/city.md>), [community](<https://devfeed.tech/tags/community.md>), [computing](<https://devfeed.tech/tags/computing.md>), [critical](<https://devfeed.tech/tags/critical.md>), [data-processing](<https://devfeed.tech/tags/data-processing.md>), [designline](<https://devfeed.tech/tags/designline.md>), [digital](<https://devfeed.tech/tags/digital.md>), [education](<https://devfeed.tech/tags/education.md>), [ee-times](<https://devfeed.tech/tags/ee-times.md>), [ee-times-university](<https://devfeed.tech/tags/ee-times-university.md>), [eelife](<https://devfeed.tech/tags/eelife.md>), [eelive-electronic-design](<https://devfeed.tech/tags/eelive-electronic-design.md>), [eet](<https://devfeed.tech/tags/eet.md>), [electrical-engineering](<https://devfeed.tech/tags/electrical-engineering.md>), [electronic-components](<https://devfeed.tech/tags/electronic-components.md>), [electronic-engineering-times](<https://devfeed.tech/tags/electronic-engineering-times.md>), [electronics](<https://devfeed.tech/tags/electronics.md>), [embedded-systems](<https://devfeed.tech/tags/embedded-systems.md>), [fundamentals-courses](<https://devfeed.tech/tags/fundamentals-courses.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [learning](<https://devfeed.tech/tags/learning.md>), [local](<https://devfeed.tech/tags/local.md>), [mixed-signal](<https://devfeed.tech/tags/mixed-signal.md>), [news](<https://devfeed.tech/tags/news.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [opinion](<https://devfeed.tech/tags/opinion.md>), [part-search](<https://devfeed.tech/tags/part-search.md>), [pilot](<https://devfeed.tech/tags/pilot.md>), [product-search](<https://devfeed.tech/tags/product-search.md>), [projects](<https://devfeed.tech/tags/projects.md>), [scewc26](<https://devfeed.tech/tags/scewc26.md>), [semiconductors](<https://devfeed.tech/tags/semiconductors.md>), [smart-cities](<https://devfeed.tech/tags/smart-cities.md>), [software](<https://devfeed.tech/tags/software.md>), [sovereign-ai](<https://devfeed.tech/tags/sovereign-ai.md>), [tech-papers](<https://devfeed.tech/tags/tech-papers.md>), [technology](<https://devfeed.tech/tags/technology.md>), [urban-infrastructure](<https://devfeed.tech/tags/urban-infrastructure.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

At the Smart City Expo, Nvidia executives and city leaders argued that municipalities should build or access local computing capacity to turn abundant civic data into operational decisions, retain control of sensitive data, and move beyond disconnected pilot projects. Examples included Doubs' data-rich digital replicas of more than 500 public buildings; Palermo's sensor-fusion system for monitoring, traffic, emergencies, and waste; Barcelona's locally hosted Laia language model for regional administrative services; West Palm Beach's public-records processing; Tampa's drone response and cross-department dashboard; and Wrocław's AI-assisted school enrollment, which handles 95% of placements. Leaders stressed public trust, human accountability, and adoption. Nvidia's proposed five-layer model spans energy, compute hardware, foundational infrastructure, software models, and applications. The suggested path is modular: start with a locally relevant, transformative use case and expand capacity over time.

### Source excerpt

City officials are moving beyond isolated pilot projects toward local computing infrastructure and AI. The post Nvidia, Municipal Leaders Frame AI as Critical Urban Infrastructure appeared first on EE Times.

## Five Ways NVIDIA Co-Design Maximizes the Value of Every Byte in the AI Factory

DevFeed: [Five Ways NVIDIA Co-Design Maximizes the Value of Every Byte in the AI Factory](<https://devfeed.tech/articles/five-ways-nvidia-co-design-maximizes-the-value-of-every-byte-in-the-ai-factory-119477.md>)

Original publisher: [Read original article](<https://developer.nvidia.com/blog/five-ways-nvidia-co-design-maximizes-the-value-of-every-byte-in-the-ai-factory/>)

Author: Shara Tibken

Published: 2026-11-05T16:00:00Z

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: [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [NVFP4](<https://devfeed.tech/topics/nvfp4.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [vllm](<https://devfeed.tech/topics/vllm.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai-generative-ai](<https://devfeed.tech/tags/agentic-ai-generative-ai.md>), [agentic-workloads](<https://devfeed.tech/tags/agentic-workloads.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai-factory](<https://devfeed.tech/tags/ai-factory.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arithmetic](<https://devfeed.tech/tags/arithmetic.md>), [attention](<https://devfeed.tech/tags/attention.md>), [batch](<https://devfeed.tech/tags/batch.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [better](<https://devfeed.tech/tags/better.md>), [bits](<https://devfeed.tech/tags/bits.md>), [block](<https://devfeed.tech/tags/block.md>), [bluefield-dpu](<https://devfeed.tech/tags/bluefield-dpu.md>), [cache](<https://devfeed.tech/tags/cache.md>), [cached-context](<https://devfeed.tech/tags/cached-context.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [co-design](<https://devfeed.tech/tags/co-design.md>), [data-center-cloud](<https://devfeed.tech/tags/data-center-cloud.md>), [doca](<https://devfeed.tech/tags/doca.md>), [dynamo](<https://devfeed.tech/tags/dynamo.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [featured](<https://devfeed.tech/tags/featured.md>), [gb200](<https://devfeed.tech/tags/gb200.md>), [gb300](<https://devfeed.tech/tags/gb300.md>), [gb300-nvl72](<https://devfeed.tech/tags/gb300-nvl72.md>), [gpudirect](<https://devfeed.tech/tags/gpudirect.md>), [h100](<https://devfeed.tech/tags/h100.md>), [hbm](<https://devfeed.tech/tags/hbm.md>), [high-bandwidth-memory](<https://devfeed.tech/tags/high-bandwidth-memory.md>), [memory](<https://devfeed.tech/tags/memory.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvlink](<https://devfeed.tech/tags/nvlink.md>), [rubin-gpu](<https://devfeed.tech/tags/rubin-gpu.md>), [spectrum-x](<https://devfeed.tech/tags/spectrum-x.md>), [tensorrt-llm](<https://devfeed.tech/tags/tensorrt-llm.md>), [vera-cpu](<https://devfeed.tech/tags/vera-cpu.md>), [vera-rubin](<https://devfeed.tech/tags/vera-rubin.md>), [vera-rubin-nvl72](<https://devfeed.tech/tags/vera-rubin-nvl72.md>)

### AI overview

The excerpt describes five ways NVIDIA's AI factory design aims to increase useful work per memory byte: NVFP4 shrinks model weights and KV cache; TensorRT-LLM Wide-EP spreads MoE experts across GPUs to free per-GPU memory and improve weight reuse; Dynamo routes requests to cached prefixes and manages KV cache across GPU, CPU, and storage; Vera CPU memory provides a place to stage context during tool-call gaps; and CMX offers pod-level shared flash for context reuse across nodes. The article reports benchmark-specific gains, including up to 4x throughput over an FP8 baseline for NVFP4 and up to 30x throughput per megawatt for Vera Rubin NVL72 versus GB300 NVL72 at the same 288 GB per GPU. It also notes that KV offloading helps when retrieval is faster than recomputation; storage performance examples are vendor validation tests.

### Source excerpt

In an AI factory, high bandwidth memory (HBM)is the fastest, most expensive, and most contested memory tier. Every byte spent on a redundant weight copy, an...

## TabPFN-3: Technical Report

DevFeed: [TabPFN-3: Technical Report](<https://devfeed.tech/articles/tabpfn-3-technical-report-116701.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.13986>)

Author: L\\'eo Grinsztajn, Klemens Fl\\"oge, Oscar Key, Felix Birkel, Philipp Jund, Brendan Roof, Mihir Manium, Shi Bin Hoo, Magnus B\\"uhler, Anurag Garg, Dominik Safaric, Jake Robertson, Benjamin J\\"ager, Simo

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

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

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [beats](<https://devfeed.tech/tags/beats.md>), [bench](<https://devfeed.tech/tags/bench.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [first](<https://devfeed.tech/tags/first.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [inference-speed](<https://devfeed.tech/tags/inference-speed.md>), [integrations](<https://devfeed.tech/tags/integrations.md>), [internet](<https://devfeed.tech/tags/internet.md>), [kv-cache](<https://devfeed.tech/tags/kv-cache.md>), [llms](<https://devfeed.tech/tags/llms.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [shap](<https://devfeed.tech/tags/shap.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [state](<https://devfeed.tech/tags/state.md>), [synthetic-data](<https://devfeed.tech/tags/synthetic-data.md>), [tabular-data](<https://devfeed.tech/tags/tabular-data.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

TabPFN-3 is a tabular foundation model pretrained exclusively on synthetic data. The report says it handles datasets up to 1 million training rows, runs up to 20 times faster than TabPFN-2.5, and, on TabArena, a single forward pass outperforms tuned and ensembled baselines while leading the speed-performance frontier. It reports gains on datasets with many classes and results above eight-hour-tuned gradient-boosted trees for datasets up to 1 million rows and 200 features. TabPFN-3-Plus (Thinking) adds test-time compute scaling; the API variant beats non-TabPFN models on TabArena and is reported 10 times faster than AutoGluon 1.5 extreme. Specialized results include TabPFN-TS-3 ranking second on fev-bench, relational-data performance on RelBenchV1, and faster SHAP computation. Reduced KV cache and row chunking enable inference on 1 million rows using one H100. These are abstract-level reported benchmark claims; the supplied text does not provide evaluation details.

### Source excerpt

arXiv:2605.13986v3 Announce Type: replace Abstract: Tabular data underpins most high-value prediction problems in science and industry, and TabPFN has driven the foundation model revolution for this modality. Designed with feedback from our users, TabPFN-3 builds on this foundation to scale state-of-the-art performance to datasets with 1M training rows and substantially reduce training and inference time. Pretrained exclusively on synthetic data from our prior, TabPFN-3 dramatically pushes the frontier of tabular prediction and brings substantial gains on time series, relational, and tabular-text data. On the standard tabular benchmark TabArena, a forward pass of TabPFN-3 outperforms all other models, including tuned and ensembled baselines, by a significant margin, and pareto-dominates the speed/performance frontier. On more diverse datasets, TabPFN-3 ranks first on datasets with many classes, and beats 8-hour-tuned gradient-boosted-tree baselines on datasets up to 1M training rows and 200 features. TabPFN-3 introduces test-time compute scaling to tabular foundation models. Our API offering TabPFN-3-Plus (Thinking) exploits this to beat all non-TabPFN models by over 200 Elo on TabArena, rising to 420 Elo on the largest data subset, and outperforms AutoGluon 1.5 extreme while being 10x faster, without using LLMs, real data, internet search or any other model besides TabPFN. TabPFN-3 extends the capabilities of our models, enabling SOTA prediction on relational data (new SOTA foundation model on RelBenchV1) and tabular-text data (SOTA on TabSTAR via TabPFN-3-Plus); and improves existing integrations: a specialized checkpoint, TabPFN-TS-3, ranks 2nd on the time-series benchmark fev-bench, and SHAP-value computation is up to 120x faster. TabPFN-3 achieves this performance while being up to 20x faster than TabPFN-2.5. In addition, a reduced KV cache and row-chunking scale to 1M rows on one H100 with fast inference speed.

## Look Before You Launch: Learned Throughput Prediction for Hardware Selection in a Production Training Fleet

DevFeed: [Look Before You Launch: Learned Throughput Prediction for Hardware Selection in a Production Training Fleet](<https://devfeed.tech/articles/look-before-you-launch-learned-throughput-prediction-for-hardware-selection-in-a-production-training-fleet-117418.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.04057>)

Author: Hossein Askari, Wei Sun, Chenguang Zhu, Bin Zan, Sreen Tallam, Keyu Man

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [model-deployment](<https://devfeed.tech/topics/model-deployment.md>)

Tags: [candidate](<https://devfeed.tech/tags/candidate.md>), [choice](<https://devfeed.tech/tags/choice.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [drivers](<https://devfeed.tech/tags/drivers.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [job](<https://devfeed.tech/tags/job.md>), [jobs](<https://devfeed.tech/tags/jobs.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [launch](<https://devfeed.tech/tags/launch.md>), [learning](<https://devfeed.tech/tags/learning.md>), [look](<https://devfeed.tech/tags/look.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [production](<https://devfeed.tech/tags/production.md>), [selection](<https://devfeed.tech/tags/selection.md>), [throughput](<https://devfeed.tech/tags/throughput.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

LEAP predicts each training job's throughput on candidate accelerator types at submission time, then ranks hardware by predicted throughput per unit cost. Using the job's training configuration, including prior throughput, it routes tens of thousands of attempts daily across hundreds of model types and more than 20 hardware types. Across 30 nightly retrains, reported accuracy was 77%, higher than the lookup table it replaced and at least as high as tested alternatives, including fine-tuned language models. Over six months, the share of eligible training GPU-hours receiving a ranking rose from 58% to 95%; the authors estimate about 3% lower training compute cost. These results are reported in the supplied abstract.

### Source excerpt

arXiv:2611.04057v1 Announce Type: new Abstract: In mixed-vendor accelerator fleets, many training jobs can run on several hardware types, and the best choice depends on each job's throughput on each type. Routing a job needs a throughput estimate for every candidate type before the job runs. We present LEAP, a machine learning performance prediction service that predicts each training job's throughput on every candidate hardware type at submission time and ranks the candidates by predicted throughput per unit cost. It routes tens of thousands of training job attempts a day across hundreds of different model types and more than 20 hardware types in a production fleet. LEAP predicts throughput from submission-time training configuration, including each job's prior throughput. LEAP reaches an accuracy of 77% over 30 nightly retrains, higher than the lookup table it replaced and at least as high as every alternative model we tested, including fine-tuned language models. Over six months, the share of eligible training GPU-hours that received a ranking by LEAP grew from 58% to 95%, and we estimate that it saves about 3% of training computational cost. In this paper, we identify the drivers that actually move accuracy and those that do not. We offer these as lessons for teams building performance predictors at fleet scale.

## Qualcomm Has Eyes On The AI Inference Datacenter With HBC And AI250

DevFeed: [Qualcomm Has Eyes On The AI Inference Datacenter With HBC And AI250](<https://devfeed.tech/articles/qualcomm-has-eyes-on-the-ai-inference-datacenter-with-hbc-and-ai250-113287.md>)

Original publisher: [Read original article](<https://www.nextplatform.com/ai/2026/11/02/qualcomm-has-eyes-on-the-ai-inference-datacenter-with-hbc-and-ai250/5306954>)

Author: Jeff Burt

Published: 2026-11-02T18:02:57Z

Content type: news

Language: en

Sources: [The Next Platform: In-depth coverage of high end computing](<https://devfeed.tech/sources/the-next-platform-in-depth-coverage-of-high-end-computing.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Qualcomm](<https://devfeed.tech/topics/qualcomm.md>), [systems](<https://devfeed.tech/topics/systems.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>)

Tags: [2-5d](<https://devfeed.tech/tags/2-5d.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-accelerator](<https://devfeed.tech/tags/ai-accelerator.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [ate](<https://devfeed.tech/tags/ate.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [chip](<https://devfeed.tech/tags/chip.md>), [chips](<https://devfeed.tech/tags/chips.md>), [compute](<https://devfeed.tech/tags/compute.md>), [computer](<https://devfeed.tech/tags/computer.md>), [computing](<https://devfeed.tech/tags/computing.md>), [datacenter](<https://devfeed.tech/tags/datacenter.md>), [hbc](<https://devfeed.tech/tags/hbc.md>), [inference](<https://devfeed.tech/tags/inference.md>), [qualcomm](<https://devfeed.tech/tags/qualcomm.md>)

### AI overview

Qualcomm says its high-bandwidth computing (HBC) design tackles AI inference's memory wall by placing an AI accelerator beneath a stack of LPDDR memory, keeping most processing near the data. The company contrasts this with on-chip model storage and HBM attached to a processor: HBC aims to reduce data movement, latency, energy use, and cost, while using a slower link to the AI250 system-on-chip for instructions and returning results. Qualcomm reports that first-generation HBC silicon passed production-level memory-path and near-memory-compute validation, delivering up to 18 times LPDDR's memory bandwidth and five times HBM4's terabits-per-second-per-watt efficiency. These are company-reported results. HBC is planned for the Dragonfly AI250 inference platform; sampling is expected in the first half of 2027. The excerpt describes validation as an early step in a multi-year roadmap and does not provide independent performance measurements.

### Source excerpt

Qualcomm executives this summer introduced an architecture called high-bandwidth computing (HBC), th ...

## How to build, maintain, and secure infrastructure for AI factories

DevFeed: [How to build, maintain, and secure infrastructure for AI factories](<https://devfeed.tech/articles/how-to-build-maintain-and-secure-infrastructure-for-ai-factories-113892.md>)

Original publisher: [Read original article](<https://canonical.com//blog/how-to-build-maintain-and-secure-infrastructure-for-ai-factories>)

Author: Ijlal Loutfi

Published: 2026-11-02T17:36:09Z

Content type: article

Language: en

Sources: [Canonical](<https://devfeed.tech/sources/canonical.md>)

Topics: [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [Ubuntu](<https://devfeed.tech/topics/ubuntu.md>), [virtual machines](<https://devfeed.tech/topics/virtual-machines.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-and-machine-learning](<https://devfeed.tech/tags/ai-and-machine-learning.md>), [ai-factories](<https://devfeed.tech/tags/ai-factories.md>), [ai-factory](<https://devfeed.tech/tags/ai-factory.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [containers](<https://devfeed.tech/tags/containers.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpu-infrastructure](<https://devfeed.tech/tags/gpu-infrastructure.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [lifecycle](<https://devfeed.tech/tags/lifecycle.md>), [maintenance](<https://devfeed.tech/tags/maintenance.md>), [models](<https://devfeed.tech/tags/models.md>), [operating-system](<https://devfeed.tech/tags/operating-system.md>), [production](<https://devfeed.tech/tags/production.md>), [secure-infrastructure](<https://devfeed.tech/tags/secure-infrastructure.md>), [security](<https://devfeed.tech/tags/security.md>), [software](<https://devfeed.tech/tags/software.md>), [support](<https://devfeed.tech/tags/support.md>)

### AI overview

Production AI infrastructure creates an ongoing maintenance obligation: GPUs may outlast several generations of models, while operating systems, containers and application dependencies have separate support lifecycles. Canonical's described approach combines Ubuntu Pro, which offers up to 15 years of OS and archive package security patching, Livepatch, hardening profile tools and FIPS-certified cryptographic modules, with maintained container builds for dependencies that host updates do not reach. Rockcraft and Chisel support building smaller OCI images; organizations still need to validate and deploy updates. Canonical Kubernetes, Charmed Kubeflow and Juju automate parts of pipeline operation and upgrades, with compatibility and performance checks before production rollout. For sensitive workloads, Ubuntu supports confidential VMs using Intel TDX or AMD SEV-SNP, including supported NVIDIA H100 configurations and remote attestation. This protects guest memory from the host hypervisor but does not replace security maintenance inside the guest.

### Source excerpt

Moving AI into production creates a long-term maintenance obligation for the software behind it. GPU infrastructure can remain in service across several generations of models and frameworks, while the operating system (OS), containers, and application dependencies each follow their own support lifecycle. Enterprise buyers need to establish who will maintain those components and how security [...]

## Europe and Canada team up in AI computing alliance that doesn't depend on Washington

DevFeed: [Europe and Canada team up in AI computing alliance that doesn't depend on Washington](<https://devfeed.tech/articles/europe-and-canada-team-up-in-ai-computing-alliance-that-doesn-t-depend-on-washington-110660.md>)

Original publisher: [Read original article](<https://www.techspot.com/news/114396-europe-canada-team-up-ai-computing-alliance-doesnt.html>)

Author: Alfonso Maruccia

Published: 2026-11-01T11:39:00Z

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>), [Supercomputing](<https://devfeed.tech/topics/supercomputing.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-factories](<https://devfeed.tech/tags/ai-factories.md>), [canada](<https://devfeed.tech/tags/canada.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [eurohpc](<https://devfeed.tech/tags/eurohpc.md>), [europe](<https://devfeed.tech/tags/europe.md>), [european-commission](<https://devfeed.tech/tags/european-commission.md>), [european-union](<https://devfeed.tech/tags/european-union.md>), [high-performance-computing-hpc](<https://devfeed.tech/tags/high-performance-computing-hpc.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [industry](<https://devfeed.tech/tags/industry.md>), [investment](<https://devfeed.tech/tags/investment.md>), [public](<https://devfeed.tech/tags/public.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [supercomputer](<https://devfeed.tech/tags/supercomputer.md>), [supercomputers](<https://devfeed.tech/tags/supercomputers.md>), [the-web](<https://devfeed.tech/tags/the-web.md>)

### AI overview

Canada is joining the EuroHPC network, giving Canadian industry, startups and research organizations access to European supercomputing resources while the EU can draw on Canada's tech industry to expand a shared AI ecosystem. EuroHPC encompasses 14 supercomputers, 10 quantum computers and 19 AI factories; the article also cites a planned EUR 30 billion investment for up to seven new AI data centers. Carnegie Endowment expert Anton Leicht describes cooperation among the EU and Canada as a practical step toward shared AI infrastructure, but warns it is unlikely to shift US dominance without massive buildouts, given US investment in gigawatt-scale clusters. The partnership is presented as part of broader EU-Canada cooperation.

### Source excerpt

European Commission President Ursula von der Leyen recently confirmed that Canada is joining Europe's network of supercomputers, which is run by the EuroHPC Joint Undertaking. Canada will become an active part of the public-private network of high-performance computing (HPC) systems, and both sides should gain something from the arrangement, at... Read Entire Article

## Chip Industry Week In Review

DevFeed: [Chip Industry Week In Review](<https://devfeed.tech/articles/chip-industry-week-in-review-107442.md>)

Original publisher: [Read original article](<https://semiengineering.com/chip-industry-week-in-review-162/>)

Author: The SE Staff

Published: 2026-10-30T07:01:27Z

Content type: news

Language: en

Sources: [Semiconductor Engineering](<https://devfeed.tech/sources/semiconductor-engineering.md>)

Topics: [Chip design](<https://devfeed.tech/topics/chip-design.md>), [tsmc](<https://devfeed.tech/topics/tsmc.md>), [intel](<https://devfeed.tech/topics/intel.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [Nemotron](<https://devfeed.tech/topics/nemotron.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [3d-integration](<https://devfeed.tech/tags/3d-integration.md>), [advanced](<https://devfeed.tech/tags/advanced.md>), [advanced-packaging](<https://devfeed.tech/tags/advanced-packaging.md>), [advantest](<https://devfeed.tech/tags/advantest.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agnisys](<https://devfeed.tech/tags/agnisys.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [alluxa](<https://devfeed.tech/tags/alluxa.md>), [altera](<https://devfeed.tech/tags/altera.md>), [amd](<https://devfeed.tech/tags/amd.md>), [amkor](<https://devfeed.tech/tags/amkor.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [applied-materials](<https://devfeed.tech/tags/applied-materials.md>), [arm](<https://devfeed.tech/tags/arm.md>), [ase](<https://devfeed.tech/tags/ase.md>), [axiomise](<https://devfeed.tech/tags/axiomise.md>), [bank-of-america](<https://devfeed.tech/tags/bank-of-america.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [blackstone](<https://devfeed.tech/tags/blackstone.md>), [bosch](<https://devfeed.tech/tags/bosch.md>), [broadcom](<https://devfeed.tech/tags/broadcom.md>), [brown-university](<https://devfeed.tech/tags/brown-university.md>), [business](<https://devfeed.tech/tags/business.md>), [business-news](<https://devfeed.tech/tags/business-news.md>), [cadence](<https://devfeed.tech/tags/cadence.md>), [chinese-academy-of-sciences](<https://devfeed.tech/tags/chinese-academy-of-sciences.md>), [chip](<https://devfeed.tech/tags/chip.md>), [chipagents](<https://devfeed.tech/tags/chipagents.md>), [citi](<https://devfeed.tech/tags/citi.md>), [classiq](<https://devfeed.tech/tags/classiq.md>), [computing-news](<https://devfeed.tech/tags/computing-news.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cxmt](<https://devfeed.tech/tags/cxmt.md>), [domain](<https://devfeed.tech/tags/domain.md>), [earnings](<https://devfeed.tech/tags/earnings.md>), [eda-news](<https://devfeed.tech/tags/eda-news.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [enpro](<https://devfeed.tech/tags/enpro.md>), [ferric](<https://devfeed.tech/tags/ferric.md>), [fraunhofer-iis](<https://devfeed.tech/tags/fraunhofer-iis.md>), [furiosaai](<https://devfeed.tech/tags/furiosaai.md>), [fz-julich](<https://devfeed.tech/tags/fz-julich.md>), [google](<https://devfeed.tech/tags/google.md>), [hardware-security-news](<https://devfeed.tech/tags/hardware-security-news.md>), [hbm](<https://devfeed.tech/tags/hbm.md>), [ibm](<https://devfeed.tech/tags/ibm.md>), [imec](<https://devfeed.tech/tags/imec.md>), [industry](<https://devfeed.tech/tags/industry.md>), [news](<https://devfeed.tech/tags/news.md>), [review](<https://devfeed.tech/tags/review.md>), [week-in-review](<https://devfeed.tech/tags/week-in-review.md>)

### AI overview

Memory and advanced-packaging capacity are emerging constraints on AI expansion: SK hynix expects shortages through 2030 and says DRAM and HBM are effectively sold out through 2029; Samsung expects tight supply through 2028 and says its 2027 HBM is sold out. TSMC is reportedly considering a 20% CoWoS price increase. The roundup also reports fab and technology investments by Infineon, UMC, Intel Foundry, Lam Research and imec, alongside new domain-specialized chip-design AI models from Cadence and an Arm framework for benchmarking CPUs on agentic workloads. Security items include a critical EDK II UEFI flaw that could enable code execution during network boot before OS protections load, and nine Supermicro BMC vulnerabilities, including a critical unauthenticated command-execution flaw. These are selected developments from the supplied roundup excerpt.

### Source excerpt

HBM shortages threaten AI expansion; TSMC weighs price hikes; advanced packaging ramps; domain-specialized AI models; Infineon and UMC boost fab investments; Intel's quantum move; Lam and imec advance 3D integration; Google expands TPUs; AI-driven earnings; CPU benchmarking methodology for agentic AI. The post Chip Industry Week In Review appeared first on Semiconductor Engineering.

## Secure AI-Scale Traffic with the Maestro MHO 200

DevFeed: [Secure AI-Scale Traffic with the Maestro MHO 200](<https://devfeed.tech/articles/secure-ai-scale-traffic-with-the-maestro-mho-200-106707.md>)

Original publisher: [Read original article](<https://blog.checkpoint.com/security/secure-ai-traffic-maestro-mho-200/>)

Author: matthewsu@checkpoint.com

Published: 2026-10-29T13:50:18Z

Content type: release

Language: en

Sources: [Check Point Blog](<https://devfeed.tech/sources/check-point-blog.md>)

Topics: [AI Networking](<https://devfeed.tech/topics/ai-networking.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [400gbe](<https://devfeed.tech/tags/400gbe.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-application](<https://devfeed.tech/tags/ai-application.md>), [ai-data-center](<https://devfeed.tech/tags/ai-data-center.md>), [ai-factories](<https://devfeed.tech/tags/ai-factories.md>), [ai-factory](<https://devfeed.tech/tags/ai-factory.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-security](<https://devfeed.tech/tags/ai-security.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [encrypted](<https://devfeed.tech/tags/encrypted.md>), [hyperscale](<https://devfeed.tech/tags/hyperscale.md>), [inference](<https://devfeed.tech/tags/inference.md>), [inspection](<https://devfeed.tech/tags/inspection.md>), [maestro](<https://devfeed.tech/tags/maestro.md>), [scale](<https://devfeed.tech/tags/scale.md>), [secure-ai](<https://devfeed.tech/tags/secure-ai.md>), [security](<https://devfeed.tech/tags/security.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

The Check Point Maestro MHO 200 is presented as a hyperscale orchestrator for inspecting rising, often encrypted data center traffic while maintaining performance. Compared with the MHO 175, it raises total fabric capacity from 3.2 to 12.8 Tbps and threat-prevention throughput per Maestro system from 1.1 to 4.5 Tbps, in the same 1RU form factor. The text also cites expanded snapshot and log storage (120 to 512 GB), remote Lights-Out Management, single-object management, and live patching without reboots or downtime. Maestro combines multiple firewalls managed and scaled as one, with failover intended to keep sessions live through failures. The article says AI applications, models, agents, and MCP servers can be discovered and secured at the firewall across Check Point firewalls; it does not provide independent performance validation. It identifies large enterprises, GPU-as-a-service providers, government AI infrastructure, telcos, and colocation operators as target users.

### Source excerpt

AI is pushing data center traffic to levels most security architectures were never built for, and much of it is encrypted. Today we're introducing the Check Point Maestro Hyperscale Orchestrator (MHO) 200, our most powerful Maestro orchestrator yet. With new 400GbE ports and four times the capacity, it lets organizations secure AI traffic without slowing it down. Why Is AI Traffic So Hard to Secure in the Data Center? Training, inference, and agentic workloads are driving up both perimeter and east-west traffic fast. A significant share of that traffic is encrypted, and all of it needs inspection, while AI services [...] The post Secure AI-Scale Traffic with the Maestro MHO 200 appeared first on Check Point Blog.

## Altera unveils Agilex 2X, betting on AI developers to expand programmable computing

DevFeed: [Altera unveils Agilex 2X, betting on AI developers to expand programmable computing](<https://devfeed.tech/articles/altera-unveils-agilex-2x-betting-on-ai-developers-to-expand-programmable-computing-105308.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/28/altera-unveils-agilex-2x-betting-on-ai-developers-to-expand-programmable-computing/>)

Author: John Furrier

Published: 2026-10-28T18:07:39Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [fpga](<https://devfeed.tech/topics/fpga.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Robotics](<https://devfeed.tech/topics/robotics.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-software](<https://devfeed.tech/tags/agentic-software.md>), [agilex](<https://devfeed.tech/tags/agilex.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-developers](<https://devfeed.tech/tags/ai-developers.md>), [altera](<https://devfeed.tech/tags/altera.md>), [applications](<https://devfeed.tech/tags/applications.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arrays](<https://devfeed.tech/tags/arrays.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [betting](<https://devfeed.tech/tags/betting.md>), [chip](<https://devfeed.tech/tags/chip.md>), [communications](<https://devfeed.tech/tags/communications.md>), [computing](<https://devfeed.tech/tags/computing.md>), [demand](<https://devfeed.tech/tags/demand.md>), [developers](<https://devfeed.tech/tags/developers.md>), [development](<https://devfeed.tech/tags/development.md>), [devices](<https://devfeed.tech/tags/devices.md>), [fpga](<https://devfeed.tech/tags/fpga.md>), [infra](<https://devfeed.tech/tags/infra.md>), [news](<https://devfeed.tech/tags/news.md>), [raghib-hussain](<https://devfeed.tech/tags/raghib-hussain.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-6](<https://devfeed.tech/tags/top-story-6.md>)

### AI overview

Altera's Agilex 2X pairs FPGA and system-on-chip families for data center and physical AI workloads with QuartusIQ tools intended to broaden access to programmable hardware. The four families target different needs, from high-throughput processing (Agilex 27) to video and sensors (Agilex 25), robotics and industrial control (Agilex 23), and input/output functions (Agilex 22). For Agilex 27, Altera attributes improved data handling to HyperFlex routing registers and the HyperFlow on-chip network, and claims 1.8 times better performance per watt than Agilex 7; customer throughput, latency, and system cost remain to be demonstrated. Quartus Prime 26.3 software and QuartusIQ are available now, while Agilex 27 engineering samples are expected in the second half of 2027. QuartusIQ connects AI agents to engineering tools and uses project context; Altera's reported speedups come from internal benchmarks, and engineers still need to verify designs, close timing, and validate hardware. The excerpt frames the opportunity as reducing the specialist effort required to build programmable systems, while noting some functions may shift to custom chips when volumes justify them.

### Source excerpt

Altera Corp. today announced Agilex 2X, a portfolio of field-programmable gate arrays and system-on-chip devices aimed at artificial intelligence infrastructure, robotics, communications and intelligent edge systems, pairing new hardware architecture with an agentic software strategy designed to accelerate development. The first family, Agilex 27, targets applications that demand high throughput and bandwidth. Altera claims a [...] The post Altera unveils Agilex 2X, betting on AI developers to expand programmable computing appeared first on SiliconANGLE.

## Inference cloud startup General Compute reels in $100M, despite broader AI funding fears

DevFeed: [Inference cloud startup General Compute reels in $100M, despite broader AI funding fears](<https://devfeed.tech/articles/inference-cloud-startup-general-compute-reels-in-100m-despite-broader-ai-funding-fears-104269.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/28/inference-cloud-startup-general-compute-reels-in-100m-despite-broader-ai-funding-fears/>)

Author: Mike Wheatley

Published: 2026-10-28T11:10:25Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Factory](<https://devfeed.tech/topics/ai-factory.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [Infrastructure as a Service (IaaS)](<https://devfeed.tech/topics/infrastructure-as-a-service-iaas.md>), [nscale](<https://devfeed.tech/topics/nscale.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-chips](<https://devfeed.tech/tags/ai-chips.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-debt](<https://devfeed.tech/tags/ai-debt.md>), [ai-funding](<https://devfeed.tech/tags/ai-funding.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-infrastructure-funding](<https://devfeed.tech/tags/ai-infrastructure-funding.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [compute](<https://devfeed.tech/tags/compute.md>), [finn-puklowski](<https://devfeed.tech/tags/finn-puklowski.md>), [funding](<https://devfeed.tech/tags/funding.md>), [general-compute](<https://devfeed.tech/tags/general-compute.md>), [inference](<https://devfeed.tech/tags/inference.md>), [inference-chips](<https://devfeed.tech/tags/inference-chips.md>), [inference-cloud](<https://devfeed.tech/tags/inference-cloud.md>), [infra](<https://devfeed.tech/tags/infra.md>), [neocloud](<https://devfeed.tech/tags/neocloud.md>), [neoclouds](<https://devfeed.tech/tags/neoclouds.md>), [news](<https://devfeed.tech/tags/news.md>), [series-a](<https://devfeed.tech/tags/series-a.md>), [startup](<https://devfeed.tech/tags/startup.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>)

### AI overview

General Compute raised $100 million in a Series A at an $800 million post-money valuation, despite warnings of consolidation among neoclouds. The company focuses on inference chips from Cerebras, SambaNova and Positron, routing customer workloads to suitable hardware; customers can use dedicated bare-metal servers or a guaranteed tokens-per-second service-level agreement. It rents space in others' data centers instead of building its own. The company also secured $400 million in debt financing in July, most of which it says remains unspent. Large-scale chip deployments are planned for next year's first quarter, with more than 75 megawatts of capacity planned for the year and one deal expected to bring over $100 million in annual recurring revenue. These are company-reported plans and expectations; the article says large-scale deployments have not yet begun.

### Source excerpt

Neoclouds are supposed to be under pressure, with the likes of Gartner Inc. and Elon Musk both predicting a bloodbath in the nascent cloud infrastructure segment in the not-too-distant future, but apparently nobody told General Compute Inc. The cloud infrastructure startup said today it has just raised $100 million in a Series A round of [...] The post Inference cloud startup General Compute reels in $100M, despite broader AI funding fears appeared first on SiliconANGLE.

[Next page](<https://devfeed.tech/topics/ai-factory.md?cursor=WyIyMDI2LTEwLTI4VDExOjEwOjI1KzAwOjAwIiwgIjQ4MGRlNDQzLTg2YWMtNGJmYy05YzQ5LTJjNmI0ZmRkYTY4NiJd>)