#parallelism

Count distinct strings with your own hash table, then on a GPU
CodingNvidia
▲ 00 solutions
Mixture of experts and expert parallelism
ML designMistral AI
▲ 00 solutions
Tensor parallelism versus pipeline parallelism
ML designMistral AIAmazon
▲ 00 solutions
Multiply two matrices across two GPUs
ML designGoogle
▲ 00 solutions
Pipeline parallelism in GRPO training
ML designGoogle
▲ 00 solutions
Data-parallel and fully sharded matrix multiply over simulated devices
CodingHardxAI
▲ 00 solutions
Dense versus mixture-of-experts at fixed compute
ML designHugging Face
▲ 00 solutions