{"version":"https://jsonfeed.org/version/1.1","title":"PyTorch — Lab Ledger Desk","home_page_url":"https://labledgerdesk.pages.dev/lab/pytorch/","feed_url":"https://labledgerdesk.pages.dev/lab/pytorch/feed.json","description":"Open-source PyTorch releases filed by Lab Ledger Desk.","items":[{"id":"https://labledgerdesk.pages.dev/b/2026/11/16/celebrating-the-2026-pytorch-foundation-contributor-award-winners-and/","url":"https://labledgerdesk.pages.dev/b/2026/11/16/celebrating-the-2026-pytorch-foundation-contributor-award-winners-and/","title":"Celebrating the 2026 PyTorch Foundation Contributor Award Winners and Nominees","content_text":"The PyTorch Foundation is proud to celebrate the winners and nominees of the fourth annual PyTorch Foundation Contributor Awards! Presented on the keynote stage at PyTorch Conference 2026 in San... The post Celebrating the 2026 PyTorch Foundation Contributor Award Winners and Nominees appeared first on PyTorch .","date_published":"2026-11-16T20:16:55.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/2026-pytorch-foundation-contributor-awards/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/16/pytorch-ambassador-community-highlights-november-2026/","url":"https://labledgerdesk.pages.dev/b/2026/11/16/pytorch-ambassador-community-highlights-november-2026/","title":"PyTorch Foundation Ambassador Community Highlights November 2026","content_text":"PyTorch Foundation Ambassadors bring people together to learn, share technical knowledge and contribute to PyTorch Foundation projects and the wider open-source AI ecosystem. In this November roundup, we highlight some... …","date_published":"2026-11-16T16:20:31.000Z","authors":[{"name":"PyTorch"}],"tags":["note","open"],"external_url":"https://pytorch.org/blog/pytorch-ambassador-community-highlights-november-2026/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/12/fast-by-default-compile-time-heuristics-for-helion-kernels/","url":"https://labledgerdesk.pages.dev/b/2026/11/12/fast-by-default-compile-time-heuristics-for-helion-kernels/","title":"Fast by Default: Compile-Time Heuristics for Helion Kernels","content_text":"TL;DR Helion now uses compile-time heuristics that analyze a kernel’s IR and dataflow to pick fast settings without slow autotuning. It classifies the kernel as matmul, reduction, or pointwise and... The post Fast by Default: Compile-Time Heuristics for Helion Kernels appeared first on PyTorch .","date_published":"2026-11-12T16:45:26.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/fast-by-default-compile-time-heuristics-for-helion-kernels/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/12/integrating-sglang-into-spyre-in-under-a-week-with-ai/","url":"https://labledgerdesk.pages.dev/b/2026/11/12/integrating-sglang-into-spyre-in-under-a-week-with-ai/","title":"Integrating SGLang into Spyre in Under a Week with AI","content_text":"TL;DR With help from AI coding agents, our two-person team integrated SGLang with IBM Spyre in less than a week, building a plugin of roughly 1,500 lines. Existing Spyre support... The post Integrating SGLang into Spyre in Under a Week with AI appeared first on PyTorch .","date_published":"2026-11-12T14:45:25.000Z","authors":[{"name":"PyTorch"}],"tags":["note","agents"],"external_url":"https://pytorch.org/blog/integrating-sglang-into-spyre-in-under-a-week-with-ai/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/12/training-a-1t-parameter-moe-on-eu-public-infrastructure-with-pytorch/","url":"https://labledgerdesk.pages.dev/b/2026/11/12/training-a-1t-parameter-moe-on-eu-public-infrastructure-with-pytorch/","title":"Training a 1T-Parameter MoE on EU public infrastructure with PyTorch and TorchTitan","content_text":"TL;DR Scaling pretraining for a ~1-trillion-parameter Mixture-of-Experts (MoE) model on shared, public infrastructure was previously considered unfeasible due to unique architectural and interconnect bottlenecks. However, by leveraging the fine-grained flexibility... …","date_published":"2026-11-12T12:00:19.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/training-a-1t-parameter-moe-on-eu-public-infrastructure-with-pytorch-and-torchtitan/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/10/executorch-webgpu-pytorch-models-on-the-gpu-in-the-browser/","url":"https://labledgerdesk.pages.dev/b/2026/11/10/executorch-webgpu-pytorch-models-on-the-gpu-in-the-browser/","title":"ExecuTorch WebGPU: PyTorch Models on the GPU, in the Browser","content_text":"TL;DR PyTorch models now run in the web browser on the GPU through ExecuTorch, with no conversion to another framework’s format. You export with torch.export to an ExecuTorch .pte file,... The post ExecuTorch WebGPU: PyTorch Models on the GPU, in the Browser appeared first on PyTorch .","date_published":"2026-11-10T22:15:07.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware"],"external_url":"https://pytorch.org/blog/executorch-webgpu-pytorch-models-on-the-gpu-in-the-browser/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/10/optimizing-muse-spark-open-on-tpu-v7-with-sglang-torchtpu/","url":"https://labledgerdesk.pages.dev/b/2026/11/10/optimizing-muse-spark-open-on-tpu-v7-with-sglang-torchtpu/","title":"Optimizing Muse Spark Open on TPU v7 with SGLang-TorchTPU","content_text":"TL;DR We serve Meta’s Muse Spark Open, an 816B-parameter mixture-of-experts model, on a single TPU v7 host of four chips, with SGLang running on the TorchTPU PyTorch backend. At the... The post Optimizing Muse Spark Open on TPU v7 with SGLang-TorchTPU appeared first on PyTorch .","date_published":"2026-11-10T16:45:24.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware"],"external_url":"https://pytorch.org/blog/optimizing-muse-spark-open-on-tpu-v7-with-sglang-torchtpu/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/04/final-call-for-proposals-pytorch-day-china-2026/","url":"https://labledgerdesk.pages.dev/b/2026/11/04/final-call-for-proposals-pytorch-day-china-2026/","title":"Call for Proposals: PyTorch Day China 2026","content_text":"The Call for Proposals for PyTorch Day China 2026 closes November 10 at 23:59 China Standard Time. The event takes place in Beijing on December 12, bringing together PyTorch developers,... The post Call for Proposals: PyTorch Day China 2026 appeared first on PyTorch .","date_published":"2026-11-04T20:29:55.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/final-call-for-proposals-pytorch-day-china-2026/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/03/kvcache-meetup-korea-2026-event-recap/","url":"https://labledgerdesk.pages.dev/b/2026/11/03/kvcache-meetup-korea-2026-event-recap/","title":"KVCache Meetup Korea 2026: Event Recap","content_text":"On the evening of Sep 30, 2026, PyTorchKR and XCENA co-hosted KVCache Meetup Korea. PyTorchKR is a major AI/ML community in Korea, and XCENA is a startup that focuses on... The post KVCache Meetup Korea 2026: Event Recap appeared first on PyTorch .","date_published":"2026-11-03T09:22:03.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/kvcache-meetup-korea-2026-event-recap/"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/02/announcing-the-special-interest-group-sig-for-ray-on-accelerators/","url":"https://labledgerdesk.pages.dev/b/2026/11/02/announcing-the-special-interest-group-sig-for-ray-on-accelerators/","title":"Announcing the Special Interest Group (SIG) for Ray on Accelerators","content_text":"Today, we are announcing the Special Interest Group (SIG) for Ray on hardware accelerators. When Ray was created in UC Berkeley over 10 years ago, the computing landscape looked very... The post Announcing the Special Interest Group (SIG) for Ray on Accelerators appeared first on PyTorch .","date_published":"2026-11-02T17:45:32.000Z","authors":[{"name":"PyTorch"}],"tags":["launch"],"external_url":"https://pytorch.org/blog/announcing-the-special-interest-group-sig-for-ray-on-accelerators/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/28/pytorch-2-15-release-blog/","url":"https://labledgerdesk.pages.dev/b/2026/10/28/pytorch-2-15-release-blog/","title":"PyTorch 2.15 Release Blog","content_text":"We are excited to announce the release of PyTorch® 2.15 (release notes)! The PyTorch 2.15 release features the following changes: Rewritten NCCL backend becomes the default, with a flight recorder... The post PyTorch 2.15 Release Blog appeared first on PyTorch .","date_published":"2026-10-28T17:44:20.000Z","authors":[{"name":"PyTorch"}],"tags":["launch"],"external_url":"https://pytorch.org/blog/pytorch-2-15-release-blog/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/26/optimizing-streaming-speech-recognition-on-arm-edge-devices-with/","url":"https://labledgerdesk.pages.dev/b/2026/10/26/optimizing-streaming-speech-recognition-on-arm-edge-devices-with/","title":"Optimizing Streaming Speech Recognition on Arm® Edge Devices with ExecuTorch","content_text":"When you’re speaking to a device, you expect text to appear as you talk. That makes streaming automatic speech recognition (ASR) a different challenge from batch transcription, where the model... The post Optimizing Streaming Speech Recognition on Arm® Edge Devices with ExecuTorch appeared first on PyTorch .","date_published":"2026-10-26T18:27:27.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/optimizing-streaming-speech-recognition-on-arm-edge-devices-with-executorch/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/22/pytorch-conference-north-america-2026-accelerating-open-intelligence/","url":"https://labledgerdesk.pages.dev/b/2026/10/22/pytorch-conference-north-america-2026-accelerating-open-intelligence/","title":"PyTorch Conference North America 2026: Accelerating Open Intelligence Together","content_text":"PyTorch Conference North America 2026 brought 3,500+ attendees to San Jose on October 20–21 for two days focused on accelerating open intelligence and exploring the impact of PyTorch Foundation projects... …","date_published":"2026-10-22T23:52:11.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/pytorch-conference-north-america-2026-accelerating-open-intelligence-together/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/22/meet-the-2026-pytorch-foundation-ambassadors/","url":"https://labledgerdesk.pages.dev/b/2026/10/22/meet-the-2026-pytorch-foundation-ambassadors/","title":"Meet the 2026 PyTorch Foundation Ambassadors","content_text":"The post Meet the 2026 PyTorch Foundation Ambassadors appeared first on PyTorch .","date_published":"2026-10-22T17:44:52.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/meet-the-2026-pytorch-foundation-ambassadors/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/21/torchtitan-update-new-sota-models-performance-at-scale-titanrl/","url":"https://labledgerdesk.pages.dev/b/2026/10/21/torchtitan-update-new-sota-models-performance-at-scale-titanrl/","title":"TorchTitan Update: New SOTA Models, Performance at Scale, TitanRL","content_text":"TorchTitan is a PyTorch-native platform for rapid experimentation and large-scale pre- and post-training of generative AI models. It is designed as a minimal, clean-room implementation of PyTorch-native techniques, with flexible... …","date_published":"2026-10-21T19:45:44.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/torchtitan-update-new-sota-models-performance-at-scale-titanrl/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/21/accelerating-pytorch-on-amd-gpus-with-flydsl/","url":"https://labledgerdesk.pages.dev/b/2026/10/21/accelerating-pytorch-on-amd-gpus-with-flydsl/","title":"Accelerating PyTorch on AMD GPUs with FlyDSL","content_text":"PyTorch users on AMD MI350-series GPUs can now use FlyDSL through existing operator APIs for dense and grouped GEMM, fused multi-head attention (FMHA/Attention), MXFP8/MXFP4 scaled GEMM, RMSNorm, and TopK, without... The post Accelerating PyTorch on AMD GPUs with FlyDSL appeared first on PyTorch .","date_published":"2026-10-21T17:45:40.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware"],"external_url":"https://pytorch.org/blog/accelerating-pytorch-on-amd-gpus-with-flydsl/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/pytorch-foundation-welcomes-nine-new-members-as-momentum-builds-around/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/pytorch-foundation-welcomes-nine-new-members-as-momentum-builds-around/","title":"PyTorch Foundation Welcomes Nine New Members As Momentum Builds Around Open Source AI Center of Gravity","content_text":"New Platinum members include Nebius, Nscale and Shopify, deepening the PyTorch Foundation’s role as the intelligence layer for open source AI development Summary The PyTorch Foundation announced nine new members,... …","date_published":"2026-10-20T16:01:06.000Z","authors":[{"name":"PyTorch"}],"tags":["launch","open"],"external_url":"https://pytorch.org/blog/pytorch-foundation-welcomes-nine-new-members-as-momentum-builds-around-open-source-ai-center-of-gravity/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/nebius-joins-the-pytorch-foundation-as-a-platinum-member/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/nebius-joins-the-pytorch-foundation-as-a-platinum-member/","title":"Nebius Joins the PyTorch Foundation as a Platinum Member","content_text":"The PyTorch Foundation, the vendor-neutral home for the open source intelligence layer, is announcing today that Nebius has joined as a Platinum member. Nebius, the AI cloud company, is building... The post Nebius Joins the PyTorch Foundation as a Platinum Member appeared first on PyTorch .","date_published":"2026-10-20T16:00:41.000Z","authors":[{"name":"PyTorch"}],"tags":["launch","open"],"external_url":"https://pytorch.org/blog/nebius-joins-the-pytorch-foundation-as-a-platinum-member/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/introducing-torchtpu/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/introducing-torchtpu/","title":"Introducing TorchTPU","content_text":"TL;DR TorchTPU is an open source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the-box model execution and TPU... The post Introducing TorchTPU appeared first on PyTorch .","date_published":"2026-10-20T16:00:30.000Z","authors":[{"name":"PyTorch"}],"tags":["launch","hardware","open"],"external_url":"https://pytorch.org/blog/introducing-torchtpu/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/from-ktir-to-spyrecode-an-mlir-dataflow-backend-for-ibm-spyre/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/from-ktir-to-spyrecode-an-mlir-dataflow-backend-for-ibm-spyre/","title":"From KTIR to SpyreCode: An MLIR Dataflow Backend for IBM Spyre","content_text":"TL;DR Torch-spyre’s Inductor decides how each kernel’s work is divided across Spyre’s cores and manages tensor residency between on-chip scratchpad and off-chip memory, and the backend compiler turns those decisions... The post From KTIR to SpyreCode: An MLIR Dataflow Backend for IBM Spyre appeared first on PyTorch .","date_published":"2026-10-20T16:00:15.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware"],"external_url":"https://pytorch.org/blog/from-ktir-to-spyrecode-an-mlir-dataflow-backend-for-ibm-spyre/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/nscale-joins-the-pytorch-foundation-as-a-platinum-member/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/nscale-joins-the-pytorch-foundation-as-a-platinum-member/","title":"Nscale Joins the PyTorch Foundation as a Platinum Member","content_text":"London, U. K., 20 October 2026 – The PyTorch Foundation, the vendor-neutral home for the open source intelligence layer, is announcing today that Nscale has joined as a Platinum member.... The post Nscale Joins the PyTorch Foundation as a Platinum Member appeared first on PyTorch .","date_published":"2026-10-20T16:00:06.000Z","authors":[{"name":"PyTorch"}],"tags":["launch","open"],"external_url":"https://pytorch.org/blog/nscale-joins-the-pytorch-foundation-as-a-platinum-member/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/19/one-search-two-layers-compiler-controls-inside-pytorch-helion-using/","url":"https://labledgerdesk.pages.dev/b/2026/10/19/one-search-two-layers-compiler-controls-inside-pytorch-helion-using/","title":"One Search, Two Layers: Compiler Controls Inside PyTorch Helion using NVIDIA CompileIQ","content_text":"With Helion’s release last year, it aimed at enabling developers to write GPU kernels in PyTorch-native Python and let its powerful autotuner do the work that used to require hand-written... The post One Search, Two Layers: Compiler Controls Inside PyTorch Helion using NVIDIA CompileIQ appeared first on PyTorch .","date_published":"2026-10-19T15:45:05.000Z","authors":[{"name":"PyTorch"}],"tags":["launch","hardware"],"external_url":"https://pytorch.org/blog/one-search-two-layers-compiler-controls-inside-pytorch-helion-using-nvidia-compileiq/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/19/introducing-torchneuron-native-pytorch-backend-for-aws-trainium/","url":"https://labledgerdesk.pages.dev/b/2026/10/19/introducing-torchneuron-native-pytorch-backend-for-aws-trainium/","title":"Introducing TorchNeuron: Native PyTorch backend for AWS Trainium","content_text":"Frontier models span language, video and audio, on architectures from dense and mixture-of-experts transformers to diffusion, state-space, world and structure-prediction models. PyTorch is where AI research happens, where the models... …","date_published":"2026-10-19T12:45:23.000Z","authors":[{"name":"PyTorch"}],"tags":["launch"],"external_url":"https://pytorch.org/blog/introducing-torchneuron-native-pytorch-backend-for-aws-trainium/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/16/bringing-elastically-deterministic-data-loading-to-torchtitan-with/","url":"https://labledgerdesk.pages.dev/b/2026/10/16/bringing-elastically-deterministic-data-loading-to-torchtitan-with/","title":"Bringing Elastically Deterministic Data Loading to TorchTitan with Zephon","content_text":"Why Zephon? The data that a model sees during training is one of the most powerful levers we have for influencing how well it performs. Much of the day-to-day work... The post Bringing Elastically Deterministic Data Loading to TorchTitan with Zephon appeared first on PyTorch .","date_published":"2026-10-16T16:30:46.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/bringing-elastically-deterministic-data-loading-to-torchtitan-with-zephon/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/15/muse-spark-1-2-is-now-open-weights/","url":"https://labledgerdesk.pages.dev/b/2026/10/15/muse-spark-1-2-is-now-open-weights/","title":"Muse Spark Is Now an Open-Weights Model","content_text":"Today, Meta is releasing the weights for Muse Spark Open, the most capable US open-weights model to date. It’s available today on Hugging Face under the Muse Model License. Muse... The post Muse Spark Is Now an Open-Weights Model appeared first on PyTorch .","date_published":"2026-10-15T17:55:52.000Z","authors":[{"name":"PyTorch"}],"tags":["launch","open"],"external_url":"https://pytorch.org/blog/muse-spark-open/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/15/bringing-executorch-to-arm-cortex-m-cpus/","url":"https://labledgerdesk.pages.dev/b/2026/10/15/bringing-executorch-to-arm-cortex-m-cpus/","title":"Bringing ExecuTorch to Arm Cortex-M CPUs","content_text":"TL;DR Deploying PyTorch models on microcontrollers has historically required converting to other framework formats, maintaining separate toolchains, and debugging cross-framework numerical mismatches. ExecuTorch eliminates this conversion step entirely. The Cortex-M... …","date_published":"2026-10-15T14:40:35.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/bringing-executorch-to-arm-cortex-m-cpus/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/13/extending-inductor-for-the-ibm-spyre-accelerator/","url":"https://labledgerdesk.pages.dev/b/2026/10/13/extending-inductor-for-the-ibm-spyre-accelerator/","title":"Extending Inductor for the IBM Spyre Accelerator","content_text":"TL;DR The PyTorch-based software stack for IBM’s Spyre accelerator recently moved its torch.compile backend from sendnn, a separate compilation stack built entirely from scratch, to backend=inductor, an open-source backend built... The post Extending Inductor for the IBM Spyre Accelerator appeared first on PyTorch .","date_published":"2026-10-13T12:45:30.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware","open"],"external_url":"https://pytorch.org/blog/extending-inductor-for-the-ibm-spyre-accelerator/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/08/session-aware-agentic-inference-with-nvidia-dynamo/","url":"https://labledgerdesk.pages.dev/b/2026/10/08/session-aware-agentic-inference-with-nvidia-dynamo/","title":"Session-Aware Agentic Inference with NVIDIA Dynamo","content_text":"TL;DR Agentic workloads change the traffic an inference server sees. Unlike single-turn chat, an agent session can involve a large initial prefill, repeated model calls, and subagents running in parallel,... The post Session-Aware Agentic Inference with NVIDIA Dynamo appeared first on PyTorch .","date_published":"2026-10-08T18:13:46.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware","agents"],"external_url":"https://pytorch.org/blog/session-aware-agentic-inference-with-nvidia-dynamo/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/08/building-spyre-as-a-native-pytorch-device/","url":"https://labledgerdesk.pages.dev/b/2026/10/08/building-spyre-as-a-native-pytorch-device/","title":"Building Spyre as a Native PyTorch Device","content_text":"TL;DR Spyre becomes a native PyTorch device by connecting PyTorch’s existing device, allocator, stream, and compiler abstractions through torch-spyre to the Spyre runtime and firmware. PrivateUse1 gives Spyre a real... The post Building Spyre as a Native PyTorch Device appeared first on PyTorch .","date_published":"2026-10-08T12:45:49.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/building-spyre-as-a-native-pytorch-device/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/06/modernizing-table-batched-embeddings-with-fbtriton/","url":"https://labledgerdesk.pages.dev/b/2026/10/06/modernizing-table-batched-embeddings-with-fbtriton/","title":"Modernizing Table Batched Embeddings with FBTriton","content_text":"This post explores the FBTriton kernel design for Table Batched Embedding (TBE) forward and backward passes. These core operators handle embedding lookups across thousands of sharded GPUs within recommendation systems.... The post Modernizing Table Batched Embeddings with FBTriton appeared first on PyTorch .","date_published":"2026-10-06T22:57:33.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware"],"external_url":"https://pytorch.org/blog/modernizing-table-batched-embeddings-with-fbtriton/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/05/evolution-of-the-pytorch-media-processing-landscape/","url":"https://labledgerdesk.pages.dev/b/2026/10/05/evolution-of-the-pytorch-media-processing-landscape/","title":"Evolution of the PyTorch Media Processing Landscape","content_text":"TL;DR If you need to decode or encode media, whether it’s images, video, or audio, use TorchCodec. If you need to transform media, use TorchVision for images and video, and... The post Evolution of the PyTorch Media Processing Landscape appeared first on PyTorch .","date_published":"2026-10-05T20:45:50.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/evolution-of-the-pytorch-media-processing-landscape/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/05/pytorch-hardware-enablement-updates-from-the-acceleration-integration/","url":"https://labledgerdesk.pages.dev/b/2026/10/05/pytorch-hardware-enablement-updates-from-the-acceleration-integration/","title":"PyTorch Hardware Enablement: Updates from the Accelerator Integration Working Group","content_text":"TL;DR The Accelerator Integration Working Group plays a vital role in standardizing how new hardware architectures connect to the open source AI ecosystem. As compute platforms diversify across cloud, edge,... …","date_published":"2026-10-05T13:12:59.000Z","authors":[{"name":"PyTorch"}],"tags":["note","hardware","open"],"external_url":"https://pytorch.org/blog/pytorch-hardware-enablement-updates-from-the-acceleration-integration-working-group/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/02/building-a-high-performance-and-portable-vllm-linear-backend-with-helion/","url":"https://labledgerdesk.pages.dev/b/2026/10/02/building-a-high-performance-and-portable-vllm-linear-backend-with-helion/","title":"Building a High-Performance and Portable vLLM Linear Backend with Helion","content_text":"TL;DR We integrated Helion into vLLM’s linear backend to explore how an autotuned, high-level kernel DSL can improve LLM inference performance while reducing kernel implementation complexity. A single Helion general... …","date_published":"2026-10-02T19:55:07.000Z","authors":[{"name":"PyTorch"}],"tags":["note","llm"],"external_url":"https://pytorch.org/blog/building-a-high-performance-and-portable-vllm-linear-backend-with-helion/"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/02/new-pathway-to-pytorch-certified-associate-ptca-certification/","url":"https://labledgerdesk.pages.dev/b/2026/10/02/new-pathway-to-pytorch-certified-associate-ptca-certification/","title":"New Pathway to PyTorch Certified Associate (PTCA) Certification","content_text":"We’re excited to bring you the new PyTorch Certified Associate (PTCA) Certification Pathway, combining focused learning modules and the PyTorch Certified Associate (PTCA) certification exam together in one structured experience... …","date_published":"2026-10-02T19:33:47.000Z","authors":[{"name":"PyTorch"}],"tags":["note"],"external_url":"https://pytorch.org/blog/new-pathway-to-pytorch-certified-associate-ptca-certification/"}]}