github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #3888] Restrict model pulling based on license #2409

New Issue
Open
opened 2026-04-12 12:43:12 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 12:43:12 -05:00
Owner
Copy Link

Originally created by @slyt on GitHub (Apr 24, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3888

It would be great if there was a configurable option to restrict the ability for Ollama server to pull certain models based on the model's license. This is necessary for organizations that would like to use Ollama as a model runtime, but cannot use some models due to limitations in their licenses.

A check could be placed after the model manifest is pulled during PullModel(). If the license in the manifest is on a block list (conversely, not on an allow list), then return an error.

Ideally the allow/block lists could be passed via CLI (e.g. --allow-license ["apache-2.0", "mit"]) or environment variable (e.g. export OLLAMA_ALLOW_LICENSE='["apache-2.0", "mit"]').

The current model manifest only contains the entire raw license text, not a simple license identifier (e.g. apache-2.0). The Licenses page on Huggingface hub docs is a good reference for common license identifiers. Some work might be required to create and track a new license identifier field in the model manifests. Hypothetically, it is possible to compare the raw license text using hashing, but requiring the user to calculate the license hash and pass it like --allow-license-hash ["12345ABCDEF"] is not user friendly.

Its likely that allowing/blocking models based on other attributes in the model manifest (hash, tag, family, etc..) would also be useful by many.

This would be a good alternative to locally hosting the model library with a subset of allowed models: https://github.com/ollama/ollama/issues/914

Originally created by @slyt on GitHub (Apr 24, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3888 It would be great if there was a configurable option to restrict the ability for Ollama server to pull certain models based on the model's license. This is necessary for organizations that would like to use Ollama as a model runtime, but cannot use some models due to limitations in their licenses. A check could be placed after [the model manifest](https://github.com/ollama/ollama/blob/790cf34d172e40c0aa4689011b5323cbdd7b83b6/server/images.go#L1034-L1037) is pulled during `PullModel()`. If the license in the manifest is on a block list (conversely, not on an allow list), then return an error. Ideally the allow/block lists could be passed via CLI (e.g. `--allow-license ["apache-2.0", "mit"]`) or environment variable (e.g. `export OLLAMA_ALLOW_LICENSE='["apache-2.0", "mit"]'`). The current model manifest only contains the entire raw license text, not a simple license identifier (e.g. `apache-2.0`). [The Licenses page on Huggingface hub docs](https://huggingface.co/docs/hub/en/repositories-licenses) is a good reference for common license identifiers. Some work might be required to create and track a new license identifier field in the model manifests. Hypothetically, it is possible to compare the raw license text using hashing, but requiring the user to calculate the license hash and pass it like `--allow-license-hash ["12345ABCDEF"]` is not user friendly. Its likely that allowing/blocking models based on other attributes in the model manifest (hash, tag, family, etc..) would also be useful by many. This would be a good alternative to locally hosting the model library with a subset of allowed models: https://github.com/ollama/ollama/issues/914
GiteaMirror added the feature request label 2026-04-12 12:43:12 -05:00
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#2409
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?