Google Cloud TPU 正式登陸 Hugging Face，支援 Inference Endpoints 與 Spaces
Original: Google Cloud TPUs made available to Hugging Face users
Hugging Face announced a deep partnership with Google Cloud, officially integrating Google Cloud TPUs (Tensor Processing Units) into the…
Hugging Face 與 Google Cloud 合作，正式將 TPU（如 TPU v5e）引入其平台。用戶現在可以在 Hugging Face Spaces 和 Inference Endpoints 中直接選擇 TPU 作為硬體加速器。這項整合為部署大型語言模型和擴散模型提供了極具性價比的 GPU 替代方案，大幅降低了雲端推理的成本與門檻。
Hugging Face announced a deep partnership with Google Cloud, officially integrating Google Cloud TPUs (Tensor Processing Units) into the Hugging Face platform. Developers can now directly select TPUs as the underlying compute infrastructure when deploying models on Hugging Face Spaces (application showcase) and Inference Endpoints.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Summaries are AI-generated; the original article is authoritative.