Announcing New vLLM and SGLang Inference Engines on TPU
First reported · 3h ago · 1 source
Crest summary — written by AI from the 1 source below; every figure is checked against them · model deepseek-flash
Summary
Google announced native PyTorch backends for the vLLM and SGLang inference engines on TPU, powered by TorchTPU.
Key facts
- The announcement was published on 2026-10-24.
- The backends are built on TorchTPU.
- The engines are vLLM and SGLang.
Why it matters
Developers running PyTorch models can use vLLM and SGLang on TPU through native backends, giving them another option for serving inference on Google's hardware.
Companies
Technologies
- PyTorchframework
- SGLangframework
- vLLMframework
- TPUproduct
- TorchTPUproduct
Related on Crest
Matched by rule, not by AI: a repository linked in the coverage, or a technology or company in this event that Crest tracks.
Repositories
- pytorch/pytorchTensors and Dynamic neural networks in Python with strong GPU accelerationMatches a named technology
- vllm-project/vllmA high-throughput and memory-efficient inference and serving engine for LLMsMatches a named technology
Packages
Original sources (1)
The report this event was built from. Crest of Tech keeps no article bodies: the full text is at each source.
- Google Developers BlogOfficial · PrimaryAnnouncing New vLLM and SGLang Inference Engines on TPUPublish time not givenRead at Google Developers Blog