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 16, 15:45 UTC
[Hardware][AWS] update neuron to 2.20 (#8676)
Signed-off-by: omrishiv <327609+omrishiv@users.noreply.github.com>
omrishiv committed
Sep 20, 2024 at 15:19 UTC
7f9c8902e3d50a9d715b38e0531280a58d2bbe14
2 files changed
+4
-4
Dockerfile.neuron
+2
-2
@@ -1,5 +1,5 @@
1
# default base image
2
-ARG BASE_IMAGE="public.ecr.aws/neuron/pytorch-inference-neuronx:2.1.2-neuronx-py310-sdk2.19.1-ubuntu20.04"
2
+ARG BASE_IMAGE="public.ecr.aws/neuron/pytorch-inference-neuronx:2.1.2-neuronx-py310-sdk2.20.0-ubuntu20.04"
3
4
FROM $BASE_IMAGE
5
@@ -20,7 +20,7 @@ RUN python3 -m pip install --upgrade pip
20
RUN python3 -m pip install --no-cache-dir fastapi ninja tokenizers pandas
21
RUN python3 -m pip install sentencepiece transformers==4.36.2 -U
22
RUN python3 -m pip install transformers-neuronx --extra-index-url=https://pip.repos.neuron.amazonaws.com -U
23
-RUN python3 -m pip install --pre neuronx-cc==2.12.* --extra-index-url=https://pip.repos.neuron.amazonaws.com -U
23
+RUN python3 -m pip install --pre neuronx-cc==2.15.* --extra-index-url=https://pip.repos.neuron.amazonaws.com -U
24
25
COPY ./vllm /app/vllm/vllm
26
COPY ./setup.py /app/vllm/setup.py
requirements-neuron.txt
+2
-2
@@ -2,6 +2,6 @@
2
-r requirements-common.txt
3
4
# Dependencies for Neuron devices
5
-transformers-neuronx >= 0.9.0
6
-torch-neuronx >= 2.1.0
5
+transformers-neuronx >= 0.12.0
6
+torch-neuronx >= 2.1.2
7
neuronx-cc