fix(--token): remove --token not exist

Dockerfile

@@ -28,14 +28,16 @@ USER myuser
 RUN mkdir -p /tmp/.cache/huggingface
 
 # to be set at runtime from secrets
-ENV TOKEN="xxx"
+ENV TASK="generate"
+ENV MAX_MODEL_LEN=32768
+ENV MAX_NUM_BATCHED_TOKENS=32768
 ENV MODEL_NAME="meta-llama/Llama-3.2-3B-Instruct"
 ENV MODEL_REVISION="0cb88a4f764b7a12671c53f0838cd831a0843b95"
 ENV HF_HOME="/tmp/.cache/huggingface"
 
 EXPOSE 7860
 
-ENTRYPOINT ["/bin/bash",  "-c", "vllm serve ${MODEL_NAME} --token ${TOKEN} --task generate --revision ${MODEL_REVISION} --code-revision ${MODEL_REVISION} --tokenizer-revision ${MODEL_NAME} --seed 42 --host 0.0.0.0 --port 7860 --max-num-batched-tokens 32768 --max-model-len 32768 --dtype float16 --enforce-eager --gpu-memory-utilization 0.9 --enable-prefix-caching --disable-log-requests --trust-remote-code"]
+ENTRYPOINT ["/bin/bash",  "-c", "vllm serve ${MODEL_NAME} --task ${TASK} --revision ${MODEL_REVISION} --code-revision ${MODEL_REVISION} --tokenizer-revision ${MODEL_NAME} --seed 42 --host 0.0.0.0 --port 7860 --max-num-batched-tokens ${MAX_NUM_BATCHED_TOKENS}} --max-model-len ${MAX_MODEL_LEN} --dtype float16 --enforce-eager --gpu-memory-utilization 0.9 --enable-prefix-caching --disable-log-requests --trust-remote-code"]
 
 # # FROM nvidia/cuda:12.1.0-cudnn8-runtime-ubuntu22.04
 # FROM nvidia/cuda:12.9.1-cudnn-runtime-ubuntu24.04