flashinfer.fused_moe.CuTileBf16Runner¶

class flashinfer.fused_moe.CuTileBf16Runner(config: MoEConfig, device: device)¶

Unified adapter for the cuTile BF16 MoE pipeline.

__init__(config: MoEConfig, device: device)¶

Methods

`__init__`(config, device)
`build`()
`check_support`()
`forward`(inputs[, tactic, do_preparation])	Forward pass for tunable runners.
`get_cache_key_extras`(inputs)	Return extra values to include in the autotune cache key.
`get_valid_tactics`(inputs, _profile)	One tactic corresponding to one cuda kernel normally, but how to interpret the meaning of tactic is pure internal details of the runner.
`pack_inputs`(act, weights)

Attributes

`backend_key`
`supported_activation_classes`
`supported_activation_classes_by_quant`
`supported_quant_variants`
`supported_routing_modes`
`supports_expert_parallelism`
`supports_fused_shared_experts`
`config`