Skip to content
KernelIndex
Search⌘K

GEMM n3072 k8192

0 eligible runs
gemm

General matrix multiply (GEMM) C = A @ B.T. Captured from Llama 3.2 3B mlp.down_proj (intermediate=8192 → hidden=3072).

Current records

No published measurement for the selected workload.

Implementations

Implementation
Runtime
Best latency
Evidence
Availability

Semantics

Inputs and outputs
afp16 [m, k]
bfp16 [n, k]
cfp16 [m, n]
Axes and behavior
kconstant = 8192
mvariable
nconstant = 3072
determinismunspecified
constraintsNo mutation or aliasing
Identity
aliasgemm_n3072_k8192modelllama-3-2-3bsha2567e9a2c9f3e13…
No source imports for this operation yet.How records are decidedJSON