For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Python class
SpecDecodeTarget
SpecDecodeTarget
class max.pipelines.speculative.spec_target.SpecDecodeTarget(*args, **kwargs)
Bases: Protocol[_BatchT]
The target model’s entry point and output layout.
ep_input_types()
ep_input_types()
Expert-parallel input types to append to the graph signature.
-
Return type:
verify()
verify(batch)
Runs the target over the merged prompt + draft tokens.
-
Parameters:
-
batch (_BatchT)
-
Return type: