For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Mojo struct
Struct_ep_dispatch_fp8
struct Struct_ep_dispatch_fp8
Implemented traits
AnyType,
ImplicitlyDestructible
Methods
execute
static def execute[input_dtype: DType, dispatch_dtype: DType, dispatch_scale_dtype: DType, hidden_size: Int, top_k: Int, n_experts: Int, max_token_per_rank: Int, n_gpus_per_node: Int, n_nodes: Int, dispatch_scale_granularity: StringSlice[StaticConstantOrigin], fused_shared_expert: Bool, skip_a2a: Bool, allreduce_world_size: Int, //, target: StringSlice[StaticConstantOrigin]](output_tokens: ManagedTensorSlice[Output, static_spec=output_tokens.static_spec], output_scales: ManagedTensorSlice[Output, static_spec=output_scales.static_spec], row_offsets: ManagedTensorSlice[Output, static_spec=row_offsets.static_spec], expert_ids: ManagedTensorSlice[Output, static_spec=expert_ids.static_spec], src_info: ManagedTensorSlice[Output, static_spec=src_info.static_spec], atomic_counters: ManagedTensorSlice[MutableInput, static_spec=atomic_counters.static_spec], input_tokens: ManagedTensorSlice[Input, static_spec=input_tokens.static_spec], topk_ids: ManagedTensorSlice[Input, static_spec=topk_ids.static_spec], send_ptrs: ManagedTensorSlice[Input, static_spec=send_ptrs.static_spec], recv_ptrs: ManagedTensorSlice[Input, static_spec=recv_ptrs.static_spec], recv_count_ptrs: ManagedTensorSlice[Input, static_spec=recv_count_ptrs.static_spec], context: DeviceContext)
Execute the fused Expert Parallelism FP8 dispatch kernel. Tokens are dispatched in Blockwise FP8 format.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!