openvino
d988c341 - [GPU] fix sdpa_micro_gqa_single_token for multi batch (#34192)

Commit
76 days ago
[GPU] fix sdpa_micro_gqa_single_token for multi batch (#34192) ### Details: - This PR fixed `sdpa_micro_gqa_single_token` to work correctly for multi batch. - updated to create an internal buffer for `sdpa_micro` for all stages including `generate` phase. - fixed the offset calculation logic for input and output in `sdpa_micro`. ### Tickets: - 181161
Author
Parents
Loading