Tensors and neural networks in Python with strong hardware acceleration. PyTorch is an open source project at the Linux Foundation. #PyTorchFoundation
Joined September 2016
- Tweets3.9K
- Following106
- Followers521K
- Likes1.2K
PyTorch retweeted
Live from the KubeCon + CloudNativeCon keynote stage: @kubeflow is now a CNCF Graduated project!
From distributed training with PyTorch and Spark Operator to the new Agents Working Group and Semantic Operator, Kubeflow bridges data, AI, and Kubernetes without vendor lock-in.
Explore the ecosystem: kubeflow.org/
#KubeCon #CloudNativeCon
PyTorch retweeted
More @PyTorch foundation projects represented on stage for the Kubecon keynotes today: @robertnishihara the creator of @raydistributed talking about the close Ray integration with @kubernetesio including Kuberay the newest @CloudNativeFdn sandbox project
PyTorch retweeted
Kevin Cochrane of @Vultr at Kubecon talking about the importance of supporting open source foundations: @aaifcommunity @PyTorch @CloudNativeFdn and @openinfradev
PyTorch retweeted
Great to hear from @ritazzhang at @CoreWeave at Kubecon about their open source contributions to @vllm_project , one of the fastest growing open source projects hosted by the @PyTorch Foundation
PyTorch retweeted
PyTorch has evolved from a tool for data scientists into a critical infrastructure bridge for modern AI technology. AI Community Architect Steve Pousty sits down with @kernelcdub to explore the @PyTorch ecosystem. bit.ly/4ekBbeb
PyTorch retweeted
Bigger models? Cool. Better systems? Even better.
Jonathan Bryce of the @LinuxFoundation hit the AGNTCon + MCPCon NA keynote stage to talk about accelerating open intelligence beyond model scaling.
Day 2 is LIVE. Tune in: bit.ly/4dVkjdO
PyTorch retweeted
vLLM 🤝 TPU
🚀 Native PyTorch serving on TPUs is here! Built with @Google and @inferact, vLLM-TorchTPU brings familiar PyTorch tools to TPU inference, with prefix caching, speculative decoding, and disaggregated serving.
🛠️ Excited to keep building with the community!
Native PyTorch on TPUs is here—and TorchTPU is officially open source.
Train and serve models using the tools you already know: @vllm_project, @sgl_project, @_llm_d_, TorchTitan, @huggingface, and @raydistributed.
Learn more from #PyTorchConf2026 → goo.gle/4hWxEVH
PyTorch retweeted
Nebius is proud to join the @PyTorch Foundation as a Platinum Member.
Ryan Hanrui Wang (@hanrui_w), our SVP Applied AI, joins the Governing Board and Technical Advisory Council.
Many teams on Nebius train and serve models with @PyTorch and @vllm_project, and we're a compute sponsor of vLLM, supporting its development, testing, and benchmarking.
We look forward to working with the community on faster training, more efficient inference, and a shorter path from research to production!
Read more: prnewswire.com/news-releases…
PyTorch retweeted
1st talk at @PyTorch: Paul Chang and Erik Schultheis on NVFP4 Training Tips and Tricks,
🗓️ Wed Oct 21 4:20 PM
What they learned training dense LLMs on GB300 at 4-bit: hyperparameters, optimisers, and recipes
Save it in the calendar: events.linuxfoundation.org/p…
PyTorch retweeted
Whether you're pushing LLM training & serving onto NPUs ⚙️, keeping accelerators in sync with PyTorch's release cycle , taming dynamic shapes in torch.compile , or exploring ultra-low-precision training.
This agenda of #PyTorchCon has you covered 👇 See you in San Jose! #PyTorch
PyTorch retweeted
Accelerating open intelligence means an open ecosystem of companies, all investing engineering time upstream to improve critical open source software for inference: @vllm_project
A story in 3 acts, last 3 days: AMD, NVIDIA, Google all contribute significant improvement to @vllm_project, measured with @SemiAnalysis_ standard benchmark on K3. Great work from @inferact team shepherding and leading many of the optimizations!!!
PyTorch retweeted
Announcing TorchNeuron: @PyTorch code now runs natively on AWS Trainium chips with a single configuration change. No rewrite needed.
Open sourcing at #PyTorchCon on Oct 20: amzn.to/3U3OnNI
PyTorch retweeted
Will be at @PyTorch conference tomorrow.
looking forward to a great line-up of speakers. 🔥
The next generation of intelligent systems won't be defined by scale.
It will be defined by intelligence that learns from your work. Co-founder @sarahookr on what that takes, live at @PyTorch Conference on October 20.
PyTorch retweeted
Day-0 support for Muse Spark Open on TokenSpeed.
190 output tokens/s per user on NVIDIA B300 — and that’s without speculative decoding.
Also supported on AMD Instinct MI355X.
👉 github.com/lightseekorg/toke…
Today we’re releasing Muse Spark Open, the most capable US open-weight model to date.
Built for coding and agentic work, it excels at long-horizon tasks, complex software engineering, and multi-step reasoning.
Key Model Specs:
→ 816B total / 42B active parameters
→ 1M token context window
→ Natively multimodal across text, image, video, audio
PyTorch retweeted
🎉 Day-0 support for Meta's open-weight Muse Spark Open is
live on vLLM, verified on NVIDIA and AMD GPUs as well as on TPUs!
816B parameters, 42B active per token, a 1M-token image, video and audio inputs, built for long-horizon agentic coding.
We’re excited to have worked with Meta Superintelligence Labs to bring Muse Spark Open to vLLM from day one as the most capable US open-weight model.
✨ 1M context for large, complex codebases
👀 Native image, video, audio & document understanding
⚡ BF16, NVFP4 & MXFP4 support
🚀 DSpark speculative decoding on both CUDA and ROCm
1/6
PyTorch retweeted
At our first #PyTorchMadrid meetup, @SergioPaniego explored TRL + OpenEnv for agent post-training!
Made me think of @goose_oss from @AgenticAIFdn 👀 Open source agent harnesses can also play a role in RL training helping models learn from tool use 💡
PyTorch retweeted
1/ 🚀 New on the vLLM blog: Finish RL Training Faster — Bitwise-Consistent, High-Performance, Agent-Driven
Same weights don't guarantee the same policy: inference optimizations can change the logprobs that training recomputes.
We make rollout and training bitwise identical while retaining high-performance inference paths. Here's how 🧵
vllm.ai/blog/2026-10-14-trai…
PyTorch retweeted
First @PyTorch Madrid meetup! On to many more!
madrid.pytorch.es
At our first #PyTorchMadrid meetup, @SergioPaniego explored TRL + OpenEnv for agent post-training!
Made me think of @goose_oss from @AgenticAIFdn 👀 Open source agent harnesses can also play a role in RL training helping models learn from tool use 💡