Safe to Delete

TensorFlow Hub Model Cache (~/.cache/tfhub_modules)

Space: 2 GB – 25 GB+ developer Windows, macOS, Linux
Quick Answer: tfhub_modules caches pre-trained neural network weights downloaded via TensorFlow Hub. Deleting it is safe.

Bottom Line: It is 100% safe to delete ~/.cache/tfhub_modules. TensorFlow Hub caches downloaded deep learning model checkpoints locally.

What is tfhub_modules?

Located at ~/.cache/tfhub_modules (or C:\tmp\tfhub_modules), Python scripts importing tensorflow_hub automatically download uncompressed model weights into this cache.

Can You Delete tfhub_modules?

✅ Yes, safe to delete.

  • Your Python training scripts will re-download model weights if executed again.

How to Delete TensorFlow Hub Cache

rm -rf ~/.cache/tfhub_modules/*

Discussion

Loading authentication...

Related in developer

The AI Engineer's Model Disk Cleanup Guide: Hugging Face, Ollama & PyTorch

The ultimate 1,200-word field guide for AI developers and local LLM hobbyists to reclaim 200GB+ from Ollama, Hugging Face, PyTorch, and vLLM model weights.

Safe to Delete

Android Studio System Index Cache (AndroidStudio Cache)

Android Studio caches Gradle project indexes, VFS caches, and compiler artifacts. Invalidate and clear is safe.

Safe to Delete

Ansible Automation Temp & Fact Cache (~/.ansible/tmp)

Ansible stores temporary playbook execution modules and target host fact JSON caches in ~/.ansible/tmp. Clearing Ansible tmp is safe.

Safe to Delete
Back to all files