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).

Python function

transition_cost

transition_cost()

max.experimental.sharding.transition_cost(source, dest, *, message_bytes, mesh, axis_index)

source

Returns the cost of redistributing source to dest on axis_index.

Dispatches on Placement.transition_to(), so custom Placement subclasses participate as long as they return one of the built-in collective names ("nop", "local_slice", "allgather", "allreduce", "reduce_scatter", "all_to_all", or "infeasible"). Anything else is reported as infeasible (+inf) so solvers reject it.

Parameters:

• source (Placement)
• dest (Placement)
• message_bytes (float)
• mesh (DeviceMesh)
• axis_index (int)

Return type:

float

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!

😔 What went wrong?

Some code doesn’t work

It includes inaccurate information

It's missing information I need

It was difficult to understand

Other

Submit