github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4555] update Ollama in a Docker container #2855

New Issue
Closed
opened 2026-04-12 13:12:11 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 13:12:11 -05:00
Owner
Copy Link

Originally created by @szRyu666 on GitHub (May 21, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4555

Originally assigned to: @dhiltgen on GitHub.

When I use Docker to run, it has already pulled the main branch and updated to the latest files. However, when I run ollama -v inside, it still shows version 0.1.33. On the other hand, using local installation and updating via curl successfully updates to version 0.1.38. How can I update Ollama running in a Docker container? I don’t want to delete the container every time and perform a complete pull operation again. Is there a way around this?

Originally created by @szRyu666 on GitHub (May 21, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4555 Originally assigned to: @dhiltgen on GitHub. When I use Docker to run, it has already pulled the main branch and updated to the latest files. However, when I run ollama -v inside, it still shows version 0.1.33. On the other hand, using local installation and updating via curl successfully updates to version 0.1.38. How can I update Ollama running in a Docker container? I don’t want to delete the container every time and perform a complete pull operation again. Is there a way around this?
GiteaMirror added the feature request label 2026-04-12 13:12:11 -05:00
GiteaMirror commented 2026-04-12 13:12:12 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (May 21, 2024):

We generate "latest" tags as well as version specific tags. https://hub.docker.com/r/ollama/ollama/tags

With Docker, if you run ollama/ollama (implying "latest") it will only pull once, and subsequent run commands will re-use that image, even if the "latest" tag has been replaced with a new version. To explicitly get the "latest", use docker pull ollama/ollama which will always check and refresh if there's a new "latest" tag, or you can pin to a specific version (e.g. ollama/ollama:0.1.38) This behavior is specific to Docker. For example, Kubernetes will always refresh the "latest" tag.

<!-- gh-comment-id:2123231646 --> @dhiltgen commented on GitHub (May 21, 2024): We generate "latest" tags as well as version specific tags. https://hub.docker.com/r/ollama/ollama/tags With Docker, if you run `ollama/ollama` (implying "latest") it will only pull once, and subsequent `run` commands will re-use that image, even if the "latest" tag has been replaced with a new version. To explicitly get the "latest", use `docker pull ollama/ollama` which will always check and refresh if there's a new "latest" tag, or you can pin to a specific version (e.g. `ollama/ollama:0.1.38`) This behavior is specific to Docker. For example, Kubernetes will always refresh the "latest" tag.
GiteaMirror commented 2026-04-12 13:12:13 -05:00
Author
Owner
Copy Link

@ArchiMickey commented on GitHub (May 22, 2024):

With Docker, if you run ollama/ollama (implying "latest") it will only pull once, and subsequent run commands will re-use that image, even if the "latest" tag has been replaced with a new version. To explicitly get the "latest", use docker pull ollama/ollama which will always check and refresh if there's a new "latest" tag, or you can pin to a specific version (e.g. ollama/ollama:0.1.38) This behavior is specific to Docker. For example, Kubernetes will always refresh the "latest" tag.

We need to create a new container for a newer image of ollama. In this case are we gonna download all the modelfiles and checkpoints again?

<!-- gh-comment-id:2124510031 --> @ArchiMickey commented on GitHub (May 22, 2024): > With Docker, if you run `ollama/ollama` (implying "latest") it will only pull once, and subsequent `run` commands will re-use that image, even if the "latest" tag has been replaced with a new version. To explicitly get the "latest", use `docker pull ollama/ollama` which will always check and refresh if there's a new "latest" tag, or you can pin to a specific version (e.g. `ollama/ollama:0.1.38`) This behavior is specific to Docker. For example, Kubernetes will always refresh the "latest" tag. We need to create a new container for a newer image of ollama. In this case are we gonna download all the modelfiles and checkpoints again?
GiteaMirror commented 2026-04-12 13:12:13 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (May 23, 2024):

We need to create a new container for a newer image of ollama. In this case are we gonna download all the modelfiles and checkpoints again?

If you follow our recommend Docker run command, the model directory is mounted from the host with -v ollama:/root/.ollama and will be preserved. If you're using a different orchestrator like Kubernetes, you'll want to set up a PVC or equivalent persistent volume strategy to ensure the model dir is preserved across container instances to avoid re-pulling the models.

<!-- gh-comment-id:2127605650 --> @dhiltgen commented on GitHub (May 23, 2024): > We need to create a new container for a newer image of ollama. In this case are we gonna download all the modelfiles and checkpoints again? If you follow our [recommend Docker run command](https://github.com/ollama/ollama/blob/main/docs/docker.md), the model directory is mounted from the host with `-v ollama:/root/.ollama` and will be preserved. If you're using a different orchestrator like Kubernetes, you'll want to set up a PVC or equivalent persistent volume strategy to ensure the model dir is preserved across container instances to avoid re-pulling the models.
GiteaMirror commented 2026-04-12 13:12:13 -05:00
Author
Owner
Copy Link

@rushasdev commented on GitHub (Apr 7, 2026):

We need to create a new container for a newer image of ollama. In this case are we gonna download all the modelfiles and checkpoints again?

If you follow our recommend Docker run command, the model directory is mounted from the host with -v ollama:/root/.ollama and will be preserved. If you're using a different orchestrator like Kubernetes, you'll want to set up a PVC or equivalent persistent volume strategy to ensure the model dir is preserved across container instances to avoid re-pulling the models.

Hi, could you update the link for the recommend Docker run command please

<!-- gh-comment-id:4200589596 --> @rushasdev commented on GitHub (Apr 7, 2026): > > We need to create a new container for a newer image of ollama. In this case are we gonna download all the modelfiles and checkpoints again? > > If you follow our [recommend Docker run command](https://github.com/ollama/ollama/blob/main/docs/docker.md), the model directory is mounted from the host with `-v ollama:/root/.ollama` and will be preserved. If you're using a different orchestrator like Kubernetes, you'll want to set up a PVC or equivalent persistent volume strategy to ensure the model dir is preserved across container instances to avoid re-pulling the models. Hi, could you update the link for the `recommend Docker run command` please
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
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#2855
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?