Results for “cpu offloading”

Get guidance on distributed training with PyTorch FSDP, which splits model parameters across devices. The skill covers implementation, mixed precision, moving data to CPU memory, communication setup, and debugging distributed training code.

• Implement distributed training with FSDP
• Review mixed precision and CPU offloading settings
• Choose a distributed communication backend

Add PyTorch FSDP2 to training scripts so model state can be split across GPUs. Covers distributed initialization, mixed precision, CPU offloading, and checkpoint saving and loading for models that exceed a single GPU's memory.

• Training a model that exceeds one GPU's memory
• Adding FSDP2 to an existing training script
• Configuring mixed precision and CPU offloading

Validate and use CPU offloading in Megatron Bridge, including layer-level activation offloading and fractional optimizer state offloading with HybridDeviceOptimizer.

The English summary is being prepared. Showing the original description.

Orientação especializada para treinamento com Fully Sharded Data Parallel do PyTorch (FSDP) - sharding de parâmetros, precisão mista, CPU offloading, FSDP2

The English summary is being prepared. Showing the original description.

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

The English summary is being prepared. Showing the original description.

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

The English summary is being prepared. Showing the original description.

Choose SGLang Diffusion settings for a specific model, GPU, and memory budget. Compare generation speed, memory use, and output quality to select command-line options, including CPU offloading, multiple GPUs, caching, and reduced precision.

• Choosing generation settings for a specific model and GPU
• Reducing GPU memory use to fit a memory budget
• Comparing performance settings across multiple GPUs

Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.

The English summary is being prepared. Showing the original description.

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

The English summary is being prepared. Showing the original description.