Files
FastDeploy/custom_ops/gpu_ops/speculate_decoding
lonelygsh e83d45833f [Speculate Decoding] Fix step_idx semantics in limit_thinking and set_stop_value kernels (#7166)
- speculate_limit_thinking_content_length: update current_base_step to
  step_idx+1 (step_idx now records history count before current round);
  remove incorrect step_idx decrement on accept_num truncation; mark
  step_idx param as const.
- speculate_set_stop_value_multi_seqs: fix can_stop gate to use
  step_idx_now+accept_num>=min_token_limit; fix skip check and pre_ids_idx
  formula (remove stale -accept_num offset); use <= condition so accept_idx
  maps directly to the accepted token that ends the stop sequence; fix
  accept_tokens index (remove -1).
- Update unit tests for speculate_set_stop_value_multi_seqs kernel.
2026-04-13 20:53:42 +08:00
..
2026-03-04 21:55:31 +08:00
2026-03-04 21:55:31 +08:00
2026-03-04 21:55:31 +08:00
2026-03-04 21:55:31 +08:00
2026-03-04 21:55:31 +08:00
2026-03-04 21:55:31 +08:00
2026-02-27 19:07:35 +08:00