// (e.g. beams[0]) as they will be removed (shifted) from all beams in all subsequent callbacks.
// These pointers are valid only during the synchronous callback, so should not be saved.
struct llama_beams_state {
- llama_beam_view * beam_views;
+ struct llama_beam_view * beam_views;
size_t n_beams; // Number of elements in beam_views[].
size_t common_prefix_length; // Current max length of prefix tokens shared by all beams.
bool last_call; // True iff this is the last callback invocation.