update SchedulerOutput

vllm/v1/core/sched/scheduler.py

@@ -157,7 +157,6 @@ class Scheduler(SchedulerInterface):
             cache_size=encoder_cache_size)
 
         speculative_config = vllm_config.speculative_config
-        # self.speculative_config = speculative_config
 
         self.use_eagle = False
         self.num_spec_tokens = self.num_lookahead_tokens = 0
@@ -1024,7 +1023,8 @@ class Scheduler(SchedulerInterface):
             # It contains the request IDs that are finished in between
             # the previous and the current steps.
             finished_req_ids=self.finished_req_ids,
-            free_encoder_input_ids=self.encoder_cache_manager.get_freed_ids(),
+            free_encoder_input_ids=self.encoder_cache_manager.
+            get_freed_mm_hashes(),
             structured_output_request_ids=structured_output_request_ids,
             grammar_bitmask=grammar_bitmask,
         )