Hugging Face Blog·· 20 h ago
Hugging Face ❤️ TorchTPU: TPUs, Natively
Hugging Face ❤️ TorchTPU: TPUs, Natively
AI summary
If you ever tried to run PyTorch on a TPU, you probably ported your code to torch_xla, or you ended up having to learn JAX. With TorchTPU, a native PyTorch backend, TPU is now just another device. It opens up native support of TPU in the Hugging Face ecosystem.
Selection record
Threshold 60Official, first-handFirst 72Second 62
AdmittedSum of both 134 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：TorchTPU原生PyTorch后端及TPU模型运行
- Why it was chosen
- Gives the exact code changes for putting a model on TPU and measured throughput figures, so the cost of migrating can be judged.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Hugging Face Blog · huggingface.co