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 13, 12:00 UTC
[Doc] Fix typo in serving docs (#28474)
Signed-off-by: the-codeboy <71213855+the-codeboy@users.noreply.github.com>
the-codeboy committed
Nov 11, 2025 at 17:45 UTC
287bbbeb067cd9e16ea9b834b35b47258a8ad43f
1 file changed
+2
-2
docs/serving/openai_compatible_server.md
+2
-2
@@ -77,11 +77,11 @@ In addition, we have the following custom APIs:
77
78
In order for the language model to support chat protocol, vLLM requires the model to include
79
a chat template in its tokenizer configuration. The chat template is a Jinja2 template that
80
-specifies how are roles, messages, and other chat-specific tokens are encoded in the input.
80
+specifies how roles, messages, and other chat-specific tokens are encoded in the input.
81
82
An example chat template for `NousResearch/Meta-Llama-3-8B-Instruct` can be found [here](https://github.com/meta-llama/llama3?tab=readme-ov-file#instruction-tuned-models)
83
84
-Some models do not provide a chat template even though they are instruction/chat fine-tuned. For those model,
84
+Some models do not provide a chat template even though they are instruction/chat fine-tuned. For those models,
85
you can manually specify their chat template in the `--chat-template` parameter with the file path to the chat
86
template, or the template in string form. Without a chat template, the server will not be able to process chat
87
and all chat requests will error.