|
liblloyal 1.0.0
Branched Inference for llama.cpp
|
Item for decode_scatter: variable tokens per branch. More...
#include <lloyal/branch.hpp>
Public Attributes | |
| BranchHandle | handle |
| std::span< const llama_token > | tokens |
Item for decode_scatter: variable tokens per branch.
Uses std::span for zero-copy non-owning view of tokens. Structural wins over parallel arrays:
tokens is a non-owning view (pointer + length). If size() > 0, data() must point to valid, dereferenceable memory.Definition at line 348 of file branch.hpp.
| BranchHandle lloyal::branch::DecodeScatterItem::handle |
Definition at line 349 of file branch.hpp.
| std::span<const llama_token> lloyal::branch::DecodeScatterItem::tokens |
Definition at line 350 of file branch.hpp.