forked from opendatahub-io/llama-stack-distribution
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathContainerfile
More file actions
70 lines (70 loc) · 2.16 KB
/
Containerfile
File metadata and controls
70 lines (70 loc) · 2.16 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
# WARNING: This file is auto-generated. Do not modify it manually.
# Generated by: distribution/build.py
FROM registry.access.redhat.com/ubi9/python-312@sha256:95ec8d3ee9f875da011639213fd254256c29bc58861ac0b11f290a291fa04435
WORKDIR /opt/app-root
RUN pip install uv
RUN pip install sqlalchemy # somehow sqlalchemy[asyncio] is not sufficient
RUN uv pip install --upgrade \
'kfp-kubernetes==2.14.6' \
'pyarrow>=21.0.0' \
'botocore==1.35.88' \
'boto3==1.35.88' \
'aiobotocore==2.16.1' \
'ibm-cos-sdk-core==2.14.2' \
'ibm-cos-sdk==2.14.2'
RUN uv pip install \
'datasets>=4.0.0' \
'mcp>=1.8.1' \
'pymilvus[milvus-lite]>=2.4.10' \
aiosqlite \
asyncpg \
autoevals \
boto3 \
chardet \
einops \
faiss-cpu \
fastapi \
fire \
google-cloud-aiplatform \
httpx \
litellm \
matplotlib \
nltk \
numpy \
opentelemetry-exporter-otlp-proto-http \
opentelemetry-sdk \
pandas \
pillow \
psycopg2-binary \
pymongo \
pypdf \
redis \
requests \
safetensors \
scikit-learn \
scipy \
sentencepiece \
sqlalchemy[asyncio] \
tokenizers \
tqdm \
transformers \
uvicorn
RUN uv pip install \
llama_stack_provider_lmeval==0.4.1
RUN uv pip install \
llama_stack_provider_ragas==0.5.1
RUN uv pip install \
llama_stack_provider_ragas[remote]==0.5.1
RUN uv pip install \
llama_stack_provider_trustyai_fms==0.3.0
RUN uv pip install --extra-index-url https://download.pytorch.org/whl/cpu 'torchao>=0.12.0' torch torchvision
RUN uv pip install --no-deps sentence-transformers
RUN uv pip install --no-cache --no-deps git+https://github.com/opendatahub-io/llama-stack.git@main
RUN mkdir -p ${HOME}/.llama ${HOME}/.cache
COPY distribution/run.yaml ${APP_ROOT}/run.yaml
COPY --chmod=755 distribution/entrypoint.sh ${APP_ROOT}/entrypoint.sh
#TODO: remove this once we have a stable version of llama-stack
# LLS server version is not aligned with the client version, so we disable the version check
# Currently, LLS client version is 0.3.0, while the server version is 0.3.0rc3+rhai0
ENV LLAMA_STACK_DISABLE_VERSION_CHECK=true
ENTRYPOINT [ "/opt/app-root/entrypoint.sh" ]