github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #9756] Check the modelfile integrity #32134

New Issue
Closed
opened 2026-04-22 13:05:40 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-22 13:05:40 -05:00
Owner
Copy Link

Originally created by @DangGwanHou on GitHub (Mar 14, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9756

what?

add new feature to ollama

why?

After migrating the model file offline, ollama will not check the integrity of the model file according to manifests, which lead to Error: model not found.

where?

when?

In the scenario of migrating model files offline.

who?

Originally created by @DangGwanHou on GitHub (Mar 14, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9756 # what? add new feature to ollama # why? After migrating the model file offline, ollama will not check the integrity of the model file according to manifests, which lead to `Error: model not found`. # where? - # when? In the scenario of migrating model files offline. # who? -
GiteaMirror added the feature requestneeds more info labels 2026-04-22 13:05:41 -05:00
GiteaMirror commented 2026-04-22 13:05:42 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 14, 2025):

Can you provide some more information about what you're requesting? Are you talking about manifests or the Modelfile? Modelfiles require you to run ollama create to build a model.

<!-- gh-comment-id:2725572700 --> @pdevine commented on GitHub (Mar 14, 2025): Can you provide some more information about what you're requesting? Are you talking about manifests or the Modelfile? Modelfiles require you to run `ollama create` to build a model.
GiteaMirror commented 2026-04-22 13:05:43 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 17, 2025):

I'm going to close this for now. We can reopen it if you add some more information.

<!-- gh-comment-id:2727750662 --> @pdevine commented on GitHub (Mar 17, 2025): I'm going to close this for now. We can reopen it if you add some more information.
GiteaMirror commented 2026-04-22 13:05:44 -05:00
Author
Owner
Copy Link

@DangGwanHou commented on GitHub (Mar 17, 2025):

@pdevine sorry for the late reply
The situation is when I try to copy modelfiles from this device(which can connect ollama model server) to other device(which cannot connect ollama model server), I ignore to check the ^sha256-.* files intergrity, then I try to run ollama ls when files copied,the command ollama ls can successfully list local models, but failed when run ollama run, the error message is that model xxx not found.
like this:

$ ollama ls
NAME                                  ID              SIZE     MODIFIED
deepseek-r1:32b-qwen-distill-fp16     7142a798da48    65 GB    2 days ago
$ ollama run deepseek-r1:32b-qwen-distill-fp16
...
Error: model "deepseek-r1:32b-qwen-distill-fp16" not found

Then I checked the files in the manifests directory and found a 32b-QWEN-Distill-FP16, and combined with the information in it, I judged that I missed copying some ^sha256-.* files.After that I supplementary copy of the missing file, and try ollama run again, the command is success.
So I wonder whether ollama will expand and supplement the function of checking the integrity of model files according to files in manifests folder in the future?

<!-- gh-comment-id:2727827726 --> @DangGwanHou commented on GitHub (Mar 17, 2025): @pdevine sorry for the late reply The situation is when I try to copy modelfiles from this device(which can connect ollama model server) to other device(which cannot connect ollama model server), I ignore to check the `^sha256-.*` files intergrity, then I try to run `ollama ls` when files copied,the command `ollama ls` can successfully list local models, but failed when run `ollama run`, the error message is that model xxx not found. like this: ```sh $ ollama ls NAME ID SIZE MODIFIED deepseek-r1:32b-qwen-distill-fp16 7142a798da48 65 GB 2 days ago $ ollama run deepseek-r1:32b-qwen-distill-fp16 ... Error: model "deepseek-r1:32b-qwen-distill-fp16" not found ``` Then I checked the files in the manifests directory and found a ` 32b-QWEN-Distill-FP16`, and combined with the information in it, I judged that I missed copying some `^sha256-.*` files.After that I supplementary copy of the missing file, and try `ollama run` again, the command is success. So I wonder whether ollama will expand and supplement the function of checking the integrity of model files according to files in manifests folder in the future?
GiteaMirror commented 2026-04-22 13:05:44 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 17, 2025):

@DangGwanHou thanks for replying. It was slightly confusing because Modelfiles are the files which you use w/ ollama create, but I think you mean the manifests and the blob files.

I'm glad you got this working. There currently isn't an ollama verify command, but it is something we've though about to check the integrity of your models. I think this will probably be more important as we break models into smaller blobs.

<!-- gh-comment-id:2728134552 --> @pdevine commented on GitHub (Mar 17, 2025): @DangGwanHou thanks for replying. It was slightly confusing because Modelfiles are the files which you use w/ `ollama create`, but I think you mean the manifests and the blob files. I'm glad you got this working. There currently isn't an `ollama verify` command, but it is something we've though about to check the integrity of your models. I think this will probably be more important as we break models into smaller blobs.
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 needs more info
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#32134
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?