Verified record · from the answer index
deepspeed
skilltested · static-verifiedscore 49/100signed passport ✓923 uses
Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention
Declared tools (1)
deepspeed
skill:orchestra-research/ai-research-skills/deepspeed · published by orchestra-research · source repository