This is the repository card of kernels-community/gpt-oss-metal-kernels that has been pushed on the Hub. It was built to be used with the kernels library. This card was automatically generated.

How to use

# make sure `kernels` is installed: `pip install -U kernels`
from kernels import get_kernel

kernel_module = get_kernel("kernels-community/gpt-oss-metal-kernels")
f32_bf16w_matmul = kernel_module.f32_bf16w_matmul

f32_bf16w_matmul(...)

Available functions

  • f32_bf16w_matmul
  • bf16_f32_embeddings
  • f32_bf16w_rmsnorm
  • f32_bf16w_dense_matmul_qkv
  • f32_bf16w_dense_matmul_attn_output
  • f32_bf16w_dense_matmul_mlp_gate
  • f32_rope
  • f32_bf16w_matmul_qkv
  • f32_sdpa
  • f32_topk
  • expert_routing_metadata
  • f32_scatter
  • f32_bf16w_matmul_add

Benchmarks

No benchmark available yet.

Downloads last month
48
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support