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 11, 03:45 UTC
[Misc] Fix spelling typos in comments (#31114)
Signed-off-by: c0de128 <kevin.mckay@outlook.com>
Kevin McKay committed
Dec 21, 2025 at 23:13 UTC
8c084de59d45a511aebe4005f49d9a81abceff3e
3 files changed
+3
-3
.buildkite/scripts/generate-nightly-index.py
+1
-1
@@ -372,7 +372,7 @@ if __name__ == "__main__":
372
373
print(f"Found {len(wheel_files)} wheel files for version {version}: {wheel_files}")
374
375
- # keep only "official" files for a non-nightly version (specifed by cli args)
375
+ # keep only "official" files for a non-nightly version (specified by cli args)
376
PY_VERSION_RE = re.compile(r"^\d+\.\d+\.\d+([a-zA-Z0-9.+-]*)?$")
377
if PY_VERSION_RE.match(version):
378
# upload-wheels.sh ensures no "dev" is in args.version
tests/models/multimodal/processing/test_tensor_schema.py
+1
-1
@@ -138,7 +138,7 @@ def create_batched_mm_kwargs(
138
)
139
140
141
-# TODO(Isotr0py): Don't initalize model during test
141
+# TODO(Isotr0py): Don't initialize model during test
142
@contextmanager
143
def initialize_dummy_model(
144
model_cls: type[nn.Module],
vllm/reasoning/mistral_reasoning_parser.py
+1
-1
@@ -104,7 +104,7 @@ class MistralReasoningParser(BaseThinkingReasoningParser):
104
# 3. Both BOT and EOT have been outputted.
105
elif has_bot_token and has_eot_token:
106
return input_ids[:bot_token_index] + input_ids[eot_token_index + 1 :]
107
- # 4. Only EOT has been outputted => this should not have occured for a model
107
+ # 4. Only EOT has been outputted => this should not have occurred for a model
108
# well prompted and trained.
109
else:
110
return input_ids[:eot_token_index] + input_ids[eot_token_index + 1 :]