biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Core][Multimodal] Track encode cache entries by mm_hash and enable embedding sharing between requests (#22711)

Browse Source 
Signed-off-by: knlnguyen1802 <knlnguyen1802@gmail.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: knlnguyen1802 <knlnguyen1802@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
This commit is contained in:
Chenguang Zheng
2025-08-25 15:41:17 +08:00
committed by GitHub
parent 712d0f88d8
commit d765cf01fe
12 changed files with 365 additions and 154 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
tests/v1/tpu/worker/test_tpu_model_runner.py
View File

@@ -85,7 +85,7 @@ def _schedule_new_request(*req_ids: str) -> SchedulerOutput:
scheduled_encoder_inputs={},
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
free_encoder_mm_hashes=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
@@ -164,7 +164,7 @@ def test_update_states_request_finished(model_runner):
scheduled_encoder_inputs={},
num_common_prefix_blocks=0,
finished_req_ids={req_id},
free_encoder_input_ids=[],
free_encoder_mm_hashes=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
@@ -194,7 +194,7 @@ def test_update_states_request_resumed(model_runner):
scheduled_encoder_inputs={},
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
free_encoder_mm_hashes=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
@@ -221,7 +221,7 @@ def test_update_states_request_resumed(model_runner):
scheduled_encoder_inputs={},
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
free_encoder_mm_hashes=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
@@ -252,7 +252,7 @@ def test_update_states_no_changes(model_runner):
scheduled_encoder_inputs={},
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
free_encoder_mm_hashes=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
@@ -287,7 +287,7 @@ def test_update_states_request_unscheduled(model_runner):
scheduled_encoder_inputs={},
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
free_encoder_mm_hashes=[],
structured_output_request_ids={},
grammar_bitmask=None,
)