https://pypi.org/project/ml-memguard/ Recent updates to the Python Package Index for ml-memguard en https://pypi.org/project/ml-memguard/0.5.0/ Stop OOM crashes in vLLM, SGLang, Unsloth, and HuggingFace. Proactive memory estimation and runtime KV-cache monitoring for LLM inference serving and fine-tuning on Apple Silicon, CUDA, and CPU. vguruprasad91@gmail.com Sat, 11 Apr 2026 13:46:34 GMT https://pypi.org/project/ml-memguard/0.1.0/ Cross-platform memory guard for ML training. Prevents OOM crashes on Apple Silicon, CUDA, and CPU by proactive estimation and runtime monitoring. vguruprasad91@gmail.com Thu, 02 Apr 2026 14:01:49 GMT