Verified record · from the answer index
distributed-llm-pretraining-torchtitan
skilltested · static-verifiedscore 49/100signed passport ✓914 uses
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distrib
Declared tools (1)
distributed-llm-pretraining-torchtitan
skill:orchestra-research/ai-research-skills/distributed-llm-pretraining-torchtitan · published by orchestra-research · source repository