NVIDIA to detail agentic workload optimization on Vera Rubin at PyTorchCon
PyTorch · x · 2026-10-21
PyTorch previews a PyTorchCon 2026 keynote: NVIDIA's Ujval Kapasi presents "PyTorch Performance for Agents on Vera Rubin," examining what agentic workloads demand from the PyTorch stack.
- New bottlenecks: a single agent request can trigger long chains of reasoning, tool calls, code execution, data movement, and distributed inference.
- Scope: hardware-software co-design across CPUs, GPUs, memory, and networking using Vera Rubin as a case study, plus software optimizations across PyTorch, Triton, and vLLM.
More from Infra
- RunInfra launches $10/mo coding plan: fastest open models across 19 agents incl. Claude Code — ycombinator · 2026-10-21
- Kimi K3 hits 391 tokens/s per user on TPU v7 with open-source vLLM megakernels — woosuk_k · 2026-10-21
- TPN Bench launches quantization service: I-matrix and quant-aware distillation for any HF model — const_reborn · 2026-10-21
- Turso unveils MVCC implementation to lift SQLite's single-writer bottleneck — glcst · 2026-10-21
- Railway pauses all new builds amid major outage, users ask if vibe coding is to blame — AaronBergman18 · 2026-10-21
- a16z: Superhuman AI hardware design is the easy part — manufacturing is the hard one — a16z · 2026-10-21