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

[V1][Core] Support for Structured Outputs (#12388)

Browse Source 
Signed-off-by: Aaron Pham <contact@aarnphm.xyz>
Signed-off-by: Russell Bryant <rbryant@redhat.com>
Co-authored-by: Russell Bryant <rbryant@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
This commit is contained in:
Aaron Pham
2025-03-07 10:19:11 -05:00
committed by GitHub
parent 1e3598edeb
commit 80e9afb5bc
26 changed files with 1528 additions and 715 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
tests/v1/worker/test_gpu_model_runner.py
View File

@@ -72,6 +72,8 @@ def _schedule_new_request(*req_ids: str) -> SchedulerOutput:
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
@@ -135,6 +137,8 @@ def test_update_states_request_finished(model_runner):
num_common_prefix_blocks=0,
finished_req_ids={req_id},
free_encoder_input_ids=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
metadata_before = model_runner.input_batch.sampling_metadata
@@ -165,6 +169,8 @@ def test_update_states_request_resumed(model_runner):
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
model_runner._update_states(scheduler_output)
@@ -190,6 +196,8 @@ def test_update_states_request_resumed(model_runner):
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
metadata_before = model_runner.input_batch.sampling_metadata
@@ -221,6 +229,8 @@ def test_update_states_no_changes(model_runner):
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
metadata_before = model_runner.input_batch.sampling_metadata
@@ -256,6 +266,8 @@ def test_update_states_request_unscheduled(model_runner):
num_common_prefix_blocks=0,
finished_req_ids=set(),
free_encoder_input_ids=[],
structured_output_request_ids={},
grammar_bitmask=None,
)
metadata_before = model_runner._update_states(scheduler_output)