Skip to content

Search Models Records GPUs Docs

More

Records GPUs Docs

Kernels Serving

⏎

Feed Serving Projects Challenges Contribute Sources Methodology Legal GitHub Atom llms.txta46b010

Kernels Serving

⏎

20 operations match

Pick one to compare its implementations

Operation

Family

Published runs

Last run

Grouped topk MoE routing backwardmoe322026-08-23 MoE expert inference batched dispatchmoe252026-08-21 MoE expert computationmoe272026-08-22 FP8 MoE expert linearmoe332026-08-23 MoE expert parallel executionmoe242026-08-21 MoE sparse routing and dispatchmoe302026-08-16 Sparse MoE routing and expertsmoe222026-08-18 MoE expert load balancing and token capacity backwardmoe222026-07-23 MoE expert batched execution with capacity factormoe242026-08-19 MoE sparse routing and dispatch backwardmoe222026-08-23 MoE group score aggregation and maskingmoe192026-07-23 MoE expert MLP with load balancingmoe282026-08-21 MoE expert parallel execution backwardmoe342026-08-21 MoE expert parallel execution with weighted aggregationmoe222026-08-20 MoE training token repeat and expert computationmoe232026-08-23 MoE expert token radix sort with prefix summoe302026-08-24 MoE sparse routing and dispatchmoe312026-08-20 MoE expert SwiGLU with down projection backwardmoe482026-08-24 MoE expert computation with weighted accumulationmoe232026-08-20 MoE layer complete forward with residualmoe302026-08-23

Ranked only against runs that measured the same thing.