We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 6b04039 commit 68af77eCopy full SHA for 68af77e
vllm/v1/worker/gpu_model_runner.py
@@ -1337,9 +1337,10 @@ def apply_grammar_bitmask(
1337
out_indices = []
1338
1339
# Reorder the bitmask to match the order of the requests in the batch.
1340
- sorted_bitmask = np.zeros_like(grammar_bitmask,
1341
- shape=(logits.shape[0],
1342
- grammar_bitmask.shape[1]))
+ sorted_bitmask = np.full(shape=(logits.shape[0],
+ grammar_bitmask.shape[1]),
+ fill_value=-1,
1343
+ dtype=grammar_bitmask.dtype)
1344
cumulative_index = 0
1345
seq = sorted(scheduler_output.structured_output_request_ids.items(),
1346
key=lambda x: x[1])
0 commit comments