Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add Dockerfile for vllm Arc support #641

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions comps/llms/text-generation/vllm/docker/Dockerfile.arc
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
# Copyright (C) 2024 Intel Corporation
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

let's remove the docker folder and rename Dockerfile.arc-> Dockerfile.intel_xpu

# SPDX-License-Identifier: Apache-2.0

FROM intelanalytics/ipex-llm-serving-vllm-xpu-experiment:2.1.0b2

COPY comps/llms/text-generation/vllm/vllm_arc.sh /llm

RUN chmod +x /llm/vllm_arc.sh

ENTRYPOINT ["/llm/vllm_arc.sh"]
18 changes: 18 additions & 0 deletions comps/llms/text-generation/vllm/vllm_arc.sh
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
#!/bin/bash

# Copyright (C) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

LLM_MODEL_ID="${LLM_MODEL_ID:=Intel/neural-chat-7b-v3-3}"

source /opt/intel/oneapi/setvars.sh
source /opt/intel/1ccl-wks/setvars.sh

python -m ipex_llm.vllm.xpu.entrypoints.openai.api_server \
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

IPEX-LLM is not yet an approved Intel product, so suggest holding on the relevant PRs for ARC support, unless the position is getting more clear.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If ipex-llm not ready enough, do we have other alternate for ARC support currently?
We can try to enable ARC with other alternates if we have.

--port 9009 \
--model ${LLM_MODEL_ID} \
--trust-remote-code \
--gpu-memory-utilization 0.9 \
--device xpu \
--enforce-eager \
$@
Loading