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.

Sign up for GitHub

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

Jump to bottom

chore: re-tag vllm worker as the default worker image #926

Merged
merged 1 commit into from
Mar 26, 2024
Merged

chore: re-tag vllm worker as the default worker image #926

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
11 changes: 11 additions & 0 deletions deploy/llms/Dockerfile.fastchat-worker
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,5 +39,16 @@ RUN python3.9 -m pip install einops scipy transformers_stream_generator==0.0.4 d
# Install requirements for Qutantization with auto-gptq
RUN python3.9 -m pip install auto-gptq optimum -i ${PYTHON_INDEX_URL}

# Install requirements for vllm worker
# Ray v2.9.3 and vllm v0.3.3
RUN python3.9 -m pip install vllm==0.3.3
RUN python3.9 -m pip install -U "ray[default]==2.9.3" -i ${PYTHON_INDEX_URL}

# Allow to use environment variable to set ray & python version to pass the version check
# for now, ray: 2.9.3, python: 3.9.x
# this utils.py file is from ray 2.9.0 ray-ml image
# search 'KubeAGI' in utils.py for what's changed
COPY deploy/llms/utils.py /usr/local/lib/python3.9/dist-packages/ray/_private/utils.py

COPY deploy/llms/start-worker.sh /
ENTRYPOINT ["/start-worker.sh"]
14 changes: 0 additions & 14 deletions deploy/llms/Dockerfile.fastchat-worker-vllm
View file
Open in desktop

This file was deleted.

5 changes: 3 additions & 2 deletions pkg/worker/runner.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,8 +33,9 @@ import (

const (
// tag is the same version as fastchat
defaultFastChatImage = "kubeagi/arcadia-fastchat-worker:v0.2.36"
defaultFastchatVLLMImage = "kubeagi/arcadia-fastchat-worker:vllm-v0.2.36"
defaultFastChatImage = "kubeagi/arcadia-fastchat-worker:v0.2.36"
// For ease of maintenance and stability, VLLM module is now included in standard image as a default feature.
defaultFastchatVLLMImage = "kubeagi/arcadia-fastchat-worker:v0.2.36"
bjwswang marked this conversation as resolved.
Show resolved Hide resolved
// defaultKubeAGIImage for RunnerKubeAGI
defaultKubeAGIImage = "kubeagi/core-library-cli:v0.0.1"
)
Expand Down
Loading