github-starred/ollama-ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2025-12-05 18:46:22 -06:00
Code Issues 2.2k Packages Projects Releases 100 Wiki Activity

do we have official ollama docker image < 1 GB #6547

New Issue
Closed
opened 2025-11-12 13:37:19 -06:00 by GiteaMirror · 8 comments
GiteaMirror commented 2025-11-12 13:37:19 -06:00
Owner
Copy Link

Originally created by @babu-kandyala on GitHub (Mar 25, 2025).

What is the issue?

i need to build custom ollama image from modelfile using ollama base image and looking for slim/lite < 1 GB or around 1 GB

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @babu-kandyala on GitHub (Mar 25, 2025). ### What is the issue? i need to build custom ollama image from modelfile using ollama base image and looking for slim/lite < 1 GB or around 1 GB ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the
bug
 label 2025-11-12 13:37:19 -06:00
GiteaMirror commented 2025-11-12 13:37:20 -06:00
Author
Owner
Copy Link

@ngxson commented on GitHub (Mar 25, 2025):

IIRC llama.cpp build size can be reduced by excluding flash attention kernels, but you will get lower performance

@ngxson commented on GitHub (Mar 25, 2025): IIRC llama.cpp build size can be reduced by excluding flash attention kernels, but you will get lower performance
GiteaMirror commented 2025-11-12 13:37:20 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 25, 2025):

The bulk of the contents is the CUDA libraries.

$ docker run --rm --entrypoint bash ollama/ollama -c 'du -sxmc /'
3302    /
3302    total
$ docker run --rm --entrypoint bash ollama/ollama -c 'du -sxmc /usr/lib/ollama/*'
1150    /usr/lib/ollama/cuda_v11
2034    /usr/lib/ollama/cuda_v12
1       /usr/lib/ollama/libggml-base.so
1       /usr/lib/ollama/libggml-cpu-alderlake.so
1       /usr/lib/ollama/libggml-cpu-haswell.so
1       /usr/lib/ollama/libggml-cpu-icelake.so
1       /usr/lib/ollama/libggml-cpu-sandybridge.so
1       /usr/lib/ollama/libggml-cpu-skylakex.so
3187    total

You could add RUN rm -rf /usr/lib/ollama/cuda_v12 to the end of the Dockerfile and free 2G, final image would be just over 1G.

@rick-github commented on GitHub (Mar 25, 2025): The bulk of the contents is the CUDA libraries. ```console $ docker run --rm --entrypoint bash ollama/ollama -c 'du -sxmc /' 3302 / 3302 total $ docker run --rm --entrypoint bash ollama/ollama -c 'du -sxmc /usr/lib/ollama/*' 1150 /usr/lib/ollama/cuda_v11 2034 /usr/lib/ollama/cuda_v12 1 /usr/lib/ollama/libggml-base.so 1 /usr/lib/ollama/libggml-cpu-alderlake.so 1 /usr/lib/ollama/libggml-cpu-haswell.so 1 /usr/lib/ollama/libggml-cpu-icelake.so 1 /usr/lib/ollama/libggml-cpu-sandybridge.so 1 /usr/lib/ollama/libggml-cpu-skylakex.so 3187 total ``` You could add `RUN rm -rf /usr/lib/ollama/cuda_v12` to the end of the Dockerfile and free 2G, final image would be just over 1G.
GiteaMirror commented 2025-11-12 13:37:20 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 25, 2025):

Actually, RUN rm will just create another layer. The files have to be excluded while building the image, not removed afterwards.

FROM ollama/ollama as base

FROM ubuntu:20.04
RUN apt-get update \
    && apt-get install -y ca-certificates \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*
COPY --from=base /bin /usr/bin
COPY --from=base /lib/ollama/*.so /usr/lib/ollama/
COPY --from=base /lib/ollama/cuda_v11 /usr/lib/ollama/cuda_v11
ENV PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
ENV NVIDIA_VISIBLE_DEVICES=all
ENV OLLAMA_HOST=0.0.0.0:11434
EXPOSE 11434
ENTRYPOINT ["/bin/ollama"]
CMD ["serve"]

$ docker build -f Dockerfile -t ollama-v11 .
$ docker image ls ollama-v11
REPOSITORY   TAG       IMAGE ID       CREATED         SIZE
ollama-v11   latest    71fece84b276   4 minutes ago   1.34GB
@rick-github commented on GitHub (Mar 25, 2025): Actually, `RUN rm` will just create another layer. The files have to be excluded while building the image, not removed afterwards. ```dockerfile FROM ollama/ollama as base FROM ubuntu:20.04 RUN apt-get update \ && apt-get install -y ca-certificates \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* COPY --from=base /bin /usr/bin COPY --from=base /lib/ollama/*.so /usr/lib/ollama/ COPY --from=base /lib/ollama/cuda_v11 /usr/lib/ollama/cuda_v11 ENV PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64 ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility ENV NVIDIA_VISIBLE_DEVICES=all ENV OLLAMA_HOST=0.0.0.0:11434 EXPOSE 11434 ENTRYPOINT ["/bin/ollama"] CMD ["serve"] ``` ```console $ docker build -f Dockerfile -t ollama-v11 . $ docker image ls ollama-v11 REPOSITORY TAG IMAGE ID CREATED SIZE ollama-v11 latest 71fece84b276 4 minutes ago 1.34GB ```
GiteaMirror commented 2025-11-12 13:37:20 -06:00
Author
Owner
Copy Link

@babu-kandyala commented on GitHub (Mar 25, 2025):

Thank you ! and also i need to create custom ollama model using the same ollama image, as per #9888 . please help

@babu-kandyala commented on GitHub (Mar 25, 2025): Thank you ! and also i need to create custom ollama model using the same ollama image, as per #9888 . please help
GiteaMirror commented 2025-11-12 13:37:21 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 25, 2025):

Add the Zscaler certificate to your image as shown in https://github.com/ollama/ollama/issues/9391#issuecomment-2698816430.

@rick-github commented on GitHub (Mar 25, 2025): Add the Zscaler certificate to your image as shown in https://github.com/ollama/ollama/issues/9391#issuecomment-2698816430.
GiteaMirror commented 2025-11-12 13:37:21 -06:00
Author
Owner
Copy Link

@babu-kandyala commented on GitHub (Mar 25, 2025):

not #9391 , i need to create custom model and make it up & running with in below dockerfile.
for Eg: ollama create custom_model -f Modelfile as per #9888

FROM ollama/ollama as base

FROM ubuntu:20.04
RUN apt-get update
&& apt-get install -y ca-certificates
&& apt-get clean
&& rm -rf /var/lib/apt/lists/*
COPY --from=base /bin /usr/bin
COPY --from=base /lib/ollama/*.so /usr/lib/ollama/
COPY --from=base /lib/ollama/cuda_v11 /usr/lib/ollama/cuda_v11
ENV PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
ENV NVIDIA_VISIBLE_DEVICES=all
ENV OLLAMA_HOST=0.0.0.0:11434
EXPOSE 11434
ENTRYPOINT ["/bin/ollama"]
CMD ["serve"]

@babu-kandyala commented on GitHub (Mar 25, 2025): not #9391 , i need to create custom model and make it up & running with in below dockerfile. for Eg: ollama create custom_model -f Modelfile as per #9888 FROM ollama/ollama as base FROM ubuntu:20.04 RUN apt-get update \ && apt-get install -y ca-certificates \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* COPY --from=base /bin /usr/bin COPY --from=base /lib/ollama/*.so /usr/lib/ollama/ COPY --from=base /lib/ollama/cuda_v11 /usr/lib/ollama/cuda_v11 ENV PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64 ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility ENV NVIDIA_VISIBLE_DEVICES=all ENV OLLAMA_HOST=0.0.0.0:11434 EXPOSE 11434 ENTRYPOINT ["/bin/ollama"] CMD ["serve"]
GiteaMirror commented 2025-11-12 13:37:21 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 25, 2025):

#9888 is because the model pull failed. You fix that by adding the Zscaler certificate to your image as shown in https://github.com/ollama/ollama/issues/9391#issuecomment-2698816430. Then you add your model as shown in https://github.com/ollama/ollama/issues/9888#issuecomment-2736224280.

@rick-github commented on GitHub (Mar 25, 2025): #9888 is because the model pull failed. You fix that by adding the Zscaler certificate to your image as shown in https://github.com/ollama/ollama/issues/9391#issuecomment-2698816430. Then you add your model as shown in https://github.com/ollama/ollama/issues/9888#issuecomment-2736224280.
GiteaMirror commented 2025-11-12 13:37:22 -06:00
Author
Owner
Copy Link

@babu-kandyala commented on GitHub (Mar 25, 2025):

will try it out, and update if any issue

@babu-kandyala commented on GitHub (Mar 25, 2025): will try it out, and update if any issue
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
client2
cloud
compatibility
context-length
create
docker
documentation
embeddings
engine
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
install
integration
intel
js
linux
macos
memory
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
python
question
registry
rendering
thinking
tools
top
windows
wsl
No Label
bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama-ollama#6547
No description provided.
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?