A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 16:00 UTC
[CI][V1] Fix passing `tokenizer` as kwarg to `validate_guidance_grammar` (#16117)
Signed-off-by: Roger Wang <ywang@roblox.com>
Roger Wang committed
Apr 6, 2025 at 01:18 UTC
9ca710e525e8db27e1a0a21649a59d208db3cf02
1 file changed
+1
-2
vllm/v1/structured_output/backend_guidance.py
+1
-2
@@ -163,7 +163,6 @@ def validate_guidance_grammar(
163
tokenizer: Optional[llguidance.LLTokenizer] = None) -> None:
164
tp, grm = get_structured_output_key(sampling_params)
165
guidance_grm = serialize_guidance_grammar(tp, grm)
166
- err = llguidance.LLMatcher.validate_grammar(guidance_grm,
167
- tokenizer=tokenizer)
166
+ err = llguidance.LLMatcher.validate_grammar(guidance_grm, tokenizer)
167
if err:
168
raise ValueError(f"Grammar error: {err}")