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, 11:00 UTC
[Misc] Slight spelling modification (#18039)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
May 13, 2025 at 11:36 UTC
c06af9a9597091d1a3ea6f893ca26a002c61e4ec
2 files changed
+2
-2
README.md
+1
-1
@@ -74,7 +74,7 @@ vLLM is flexible and easy to use with:
74
- OpenAI-compatible API server
75
- Support NVIDIA GPUs, AMD CPUs and GPUs, Intel CPUs and GPUs, PowerPC CPUs, TPU, and AWS Neuron.
76
- Prefix caching support
77
-- Multi-lora support
77
+- Multi-LoRA support
78
79
vLLM seamlessly supports most popular open-source models on HuggingFace, including:
80
- Transformer-like LLMs (e.g., Llama)
vllm/model_executor/layers/logits_processor.py
+1
-1
@@ -119,7 +119,7 @@ class LogitsProcessor(nn.Module):
119
120
def extra_repr(self) -> str:
121
s = f"vocab_size={self.vocab_size}"
122
- s += f", forg_vocab_size={self.org_vocab_size}"
122
+ s += f", org_vocab_size={self.org_vocab_size}"
123
s += f", scale={self.scale}, logits_as_input={self.logits_as_input}"
124
return s
125