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

⏎

dtype bf16✕

20 operations match

Pick one to compare its implementations; your filters carry over

Operation

Family

Published runs

Last run

FP8 MoE expert linearnone on bf16moe332026-08-23 Grouped topk MoE routing backwardnone on bf16moe322026-08-23 MoE expert batched execution with capacity factornone on bf16moe242026-08-19 MoE expert computationnone on bf16moe272026-08-22 MoE expert computation with weighted accumulationnone on bf16moe232026-08-20 MoE expert inference batched dispatchnone on bf16moe252026-08-21 MoE expert load balancing and token capacity backwardnone on bf16moe222026-07-23 MoE expert MLP with load balancingnone on bf16moe282026-08-21 MoE expert parallel executionnone on bf16moe242026-08-21 MoE expert parallel execution backwardnone on bf16moe342026-08-21 MoE expert parallel execution with weighted aggregationnone on bf16moe222026-08-20 MoE expert SwiGLU with down projection backwardnone on bf16moe482026-08-24 MoE expert token radix sort with prefix sumnone on bf16moe302026-08-24 MoE group score aggregation and maskingnone on bf16moe192026-07-23 MoE layer complete forward with residualnone on bf16moe302026-08-23 MoE sparse routing and dispatchnone on bf16moe302026-08-16 MoE sparse routing and dispatchnone on bf16moe312026-08-20 MoE sparse routing and dispatch backwardnone on bf16moe222026-08-23 MoE training token repeat and expert computationnone on bf16moe232026-08-23 Sparse MoE routing and expertsnone on bf16moe222026-08-18

Ranked only against runs that measured the same thing.