# Driving the Future of Open Source AI: An Update from PyTorch Foundation Projects > Source: > Published: 2026-07-22 14:44:38+00:00 ### Featured projects **TL;DR** In April 2025, the PyTorch Foundation evolved into a [multi-project Foundation](https://pytorch.org/blog/pt-foundation-expands/), with the objective to support deeper collaboration across domains and help scale innovation throughout the AI lifecycle. Today, the Foundation hosts six projects, including PyTorch, vLLM, DeepSpeed, Ray, Helion, and Safetensors. From core optimization updates to hardware enablement and community health, PyTorch Foundation-hosted projects have achieved a lot this past quarter. In this new blog series, each of our projects provides an update on its latest achievements. ## PyTorch Updates: 2.13 Release, ExecuTorch, and Apple Silicon Optimization Development momentum for the core PyTorch project remains incredibly strong. Commits continued to rise with 4,415 in Q2 alone, and issue management is on track, with the overall number of open issues steadily decreasing. The community officially rolled out [PyTorch 2.13](https://pytorch.org/blog/pytorch-2-13-release-blog/?_gl=1*1ozwfbk*_up*MQ..*_ga*OTU1Njk5NDY1LjE3ODQyMzYwMjU.*_ga_469Y0W5V62*czE3ODQyMzYwMjQkbzEkZzEkdDE3ODQyMzY0MjQkajYwJGwwJGgw) and has already kicked off the development cycle for PyTorch 2.14. PyTorch 2.13 headlines performance and platform reach: FlexAttention arrives on Apple Silicon (up to ~12x faster than SDPA) alongside a new CuTeDSL Inductor backend, a memory-saving nn.LinearCrossEntropyLoss (up to 4x less peak GPU memory), and Python 3.15 wheel support including free-threaded builds. On the distributed and platform side, it adds the torchcomms backend and FSDP2 communication overlap for large-cluster training, broadens ROCm/Arm/Intel XPU support. The team is actively expanding on-device LLM capabilities through ExecuTorch, releasing weekly examples alongside a partnership with Hugging Face to streamline model ingestion and support on-device performance. ## vLLM Updates: Model Runner V2, Q3 2026 Roadmap, and vLLM Conference The vLLM project has reached an impressive milestone, achieving a stable, bi-weekly release cadence. Among its key achievements is the complete redesign of Model Runner V2, which yielded substantial performance improvements on GPTQ. vLLM continues to offer day-zero support for massive, state-of-the-art models such as Kimi K3, Minimax M3, and Qwen 3.8. Looking ahead, the team has published its [Q3 2026 roadmap](https://github.com/vllm-project/vllm/issues/48168). The roadmap is organized across special interest groups (SIGs) and centers on production agentic workloads and high-interactivity premium tokens. On the core engine, the team is finishing two major migrations, Flat Model and Model Runner V2, and redesigning how the scheduler and KV cache work. For large-scale serving, the goal is top performance on AgentX, with better KV cache offloading and smarter prefix caching for multi-turn agents. Other groups are focused on faster speculative decoding, production-ready KV-cache compression, and quicker CI. There are also dedicated tracks for model performance, vLLM-Omni, and reinforcement learning. The team is also hosting the first-ever [vLLM Conference](https://vllm.ai/events/vllm-conference/2026), held at Ray Summit in San Francisco from August 24–26. The event brings together engineers and researchers from across the vLLM ecosystem for two days of sessions spanning the vLLM roadmap, hardware backends, agentic serving, training, and production-scale inference. ## DeepSpeed Updates: Ulysses Parallelism, Torch.xpu Integration, and ASPLOS 2026 DeepSpeed continues to deliver rapid software innovations, maintaining its stable bi-weekly schedule with six software releases in the first quarter. To foster closer collaboration with developers, the team stabilized its monthly virtual office hours as consistent community touchpoints. In a major architectural update, the project replaced Intel GPU support (IPEX) by directly integrating Torch.xpu. The team also rolled out updates focusing on mixed precision fixes, AMD SDMA optimization, automatic sequence parallelism, and Muon optimizer. One of the main highlights for the project is the integration of the Ulysses sequence parallelism algorithm directly into Hugging Face libraries, including Trainer, Accelerate, and TRL. DeepSpeed’s theoretical contributions have also gained recognition in the academic community, earning a best-paper honorable mention at [ASPLOS 2026](https://www.asplos-conference.org/asplos2026/awards/index.html) for its pioneering [SuperOffload work](https://arxiv.org/abs/2509.21271), and having three research papers accepted. The first paper (AutoSP) [automates sequence parallelism using PyTorch 2.0 compile technology](https://iclr.cc/virtual/2026/poster/10011914), the second paper explores [energy analysis and offloading overheads on Grace Hopper C2C systems](https://arxiv.org/abs/2605.01938), while the third paper focuses on [efficient offloading mechanics for model training](https://arxiv.org/abs/2505.12242). ## Ray Updates: GB200/GB300 Hardware Support, Frontier Model Scaling, and Ray Data 2.57 Ray remains focused on production hardening for large-scale workloads. This work includes speeding up actor scheduling, upgrading native RDMA support, improving topology-aware scheduling for rack-aware actor placement, and improving support for GB200s/GB300s and beyond. The team is also focused on optimizing workflows for reinforcement learning as well as for multimodal data and video processing. In the last few months, Ray has been used for building a variety of frontier models, including [MAI-Thinking-1](https://microsoft.ai/pdf/mai-thinking-1.pdf), [Composer 2.5](https://arxiv.org/pdf/2603.24477), and [Nemotron 3 Ultra](https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Ultra-Technical-Report.pdf). A few other recent highlights from the team include [inference performance improvements](https://cloud.google.com/blog/products/containers-kubernetes/improving-ray-serve-llm-on-gke-throughput-latency) with Ray + vLLM, improved cluster stability via [resource isolation](https://www.anyscale.com/blog/enhancing-ray-cluster-stability-with-resource-isolation), improved [data pipeline reliability](https://www.anyscale.com/blog/ray-data-256-updates), and prefill-decode disaggregation with [Ray + vLLM on AMD MI325X](https://www.anyscale.com/blog/ray-vllm-prefill-decode-disaggregation-amd-mi325x-67-percent-savings). The team has also made important updates to data and inference pipelines. A brand-new, high-performance engine for Ray Data is scheduled to debut in version 2.57. At the same time, the team is actively building out weight-syncing integrations with vLLM and SkyRL to eliminate performance bottlenecks during post-training. If you want to connect with the team and community in person, [Ray Summit](https://www.anyscale.com/ray-summit/2026) is set for August 24-26 in San Francisco. ## Helion Updates: Cross-Hardware Attention Kernels and LLM-Guided Autotuning Helion has made substantial progress during its first six months of the year, focusing its efforts on delivering state-of-the-art performance across heterogeneous hardware. The project introduced the CuTeDSL backend for NVIDIA GPUs and Pallas backend for TPUs. With the new CuteDSL and Pallas backends, using [the same Helion attention kernel](https://github.com/pytorch/helion/blob/main/examples/attention.py), Helion delivers state-of-the-art performance on NVIDIA Blackwell GPU, outperforming FlashAttention-4, and on Google TPU, outperforming hand-written Attention kernel from the [Tokamax library](https://github.com/openxla/tokamax), demonstrating Helion’s performance portability. The Pallas backend is being built in partnership with Google and debuted at the [Google AI Systems DevLabs](https://rsvp.withgoogle.com/events/dev-lab-summer-2026). In addition to performance improvements and support for new hardware, Helion incorporated the power of LLMs to kernel autotuning. The team recently introduced an LLM-guided autotuner to significantly speedup autotuning, demonstrating an [10x improvement in tuning efficiency](https://pytorch.org/blog/from-minutes-to-seconds-llm-guided-autotuning-for-helion-kernels/), paving the way for faster, highly optimized deployments. To engage with the community, Helion held its [first hackathon](https://cerebralvalley.ai/e/helion-hackathon) in partnership with NVIDIA and GPU Mode and a tutorial at the [PLDI conference](https://pldi26.sigplan.org/details/pldi-2026-tutorials/1/Writing-Performance-Portable-Kernels-Simplified-with-Helion). Helion was announced in [General Availability](https://pytorchconferenceeu2026.sched.com/event/2Himi/helion-10-a-high-level-dsl-for-performance-portable-kernels-oguz-ulgen-meta) at the PyTorch Conference Europe. ## Safetensors Updates: GIL-Free Serialization, Python 3.14 Support, and MPS Fast-Loading In June, Safetensors officially rolled out Global Interpreter Lock (GIL)-Free Serialization, which introduces enhanced parallel processing capabilities to improve overall multi-threaded performance. To provide early support for upcoming Python releases, the project has expanded its continuous integration pipeline to accommodate Python 3.14 and 3.14t. For developers operating on compatible hardware, the new Metal Performance Shaders (MPS) fast-load paths introduce native Metal buffers within the torch MPS backend to significantly minimize model loading latency. Finally, new dtypes have been added to offer developers greater precision and flexibility across diverse modeling workloads. Looking ahead, Safetensors is actively working on optimizing CUDA fast-path loading capabilities, exploring integrations with both io_uring and GPU Direct Storage (GDS). As an initial step, the project is implementing multithreaded pread operations into pinned host buffers to accelerate data transfers. This structural work aims to make the library significantly more adaptable to additional custom fast paths while removing our historical reliance on framework-managed storage, such as the from_file memory mapping found in torch. The team is currently evaluating the feasibility of this transition, as it depends on DLPack handoffs that may not be universally supported across all frameworks. ## Join the Community Are you ready to build the future of AI with us? To learn more about our hosted projects, visit our [Foundation-Hosted Projects portal](https://pytorch.org/projects/), and to learn how your organization can become a member, visit our [Join Now](https://pytorch.org/join/) page. Plus, join us in Shanghai for [KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026](https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/) September 7-9, and in San Jose for [PyTorch Conference North America 2026](https://events.linuxfoundation.org/pytorch-conference-north-america/), October 20-21, later this year.