Skip to content
KernelIndex
Search⌘K

GEMM n4608 k3584

0 eligible runs
gemm

General matrix multiply (GEMM) C = A @ B.T. Captured from Qwen2.5-7B attn.qkv_proj ((28+4+4)*128 = 4608, hidden=3584).

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 = 3584
mvariable
nconstant = 4608
determinismunspecified
constraintsNo mutation or aliasing
Identity
aliasgemm_n4608_k3584modelqwen2-5-7bsha256b982403d2f20…
No source imports for this operation yet.How records are decidedJSON