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 #3185] ollama doesn't distribute notice licenses in its release artifacts #1963

New Issue
Open
opened 2026-04-12 12:07:23 -05:00 by GiteaMirror · 50 comments
GiteaMirror commented 2026-04-12 12:07:23 -05:00
Owner
Copy Link

Originally created by @jart on GitHub (Mar 16, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3185

Originally assigned to: @BruceMacD on GitHub.

What is the issue?

ollama uses projects like llama.cpp as a statically linked dependency. The terms of the MIT license require that it distribute the copyright notice in both source and binary form. Yet if I grep for "Georgi Gerganov" on my Linux and Windows installation folders for ollama, the copyright notices are nowhere to be found. This is a violation of the terms of the license which should be rectified.

What did you expect to see?

I expected the copyright notices of dependent projects to be at minimum present within the binary artifacts. Some people online are of the opinion that a mention of llama.cpp should be present in places like the README as well, although the license doesn't strictly require that.

Steps to reproduce

No response

Are there any recent changes that introduced the issue?

grep -iR 'Georgi Gerganov' AppData/Local/Programs/Ollama/
grep -R 'Georgi Gerganov' /usr/local/bin/ollama
etc.

OS

Other

Architecture

Other

Platform

No response

Ollama version

No response

GPU

No response

GPU info

No response

CPU

No response

Other software

No response

Originally created by @jart on GitHub (Mar 16, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3185 Originally assigned to: @BruceMacD on GitHub. ### What is the issue? ollama uses projects like llama.cpp as a statically linked dependency. The terms of the MIT license require that it distribute the copyright notice in both source and binary form. Yet if I `grep` for "Georgi Gerganov" on my Linux and Windows installation folders for ollama, the copyright notices are nowhere to be found. This is a violation of the terms of the license which should be rectified. ### What did you expect to see? I expected the copyright notices of dependent projects to be at minimum present within the binary artifacts. Some people online are of the opinion that a mention of llama.cpp should be present in places like the README as well, although the license doesn't strictly require that. ### Steps to reproduce _No response_ ### Are there any recent changes that introduced the issue? ``` grep -iR 'Georgi Gerganov' AppData/Local/Programs/Ollama/ grep -R 'Georgi Gerganov' /usr/local/bin/ollama etc. ``` ### OS Other ### Architecture Other ### Platform _No response_ ### Ollama version _No response_ ### GPU _No response_ ### GPU info _No response_ ### CPU _No response_ ### Other software _No response_
GiteaMirror added the bug label 2026-04-12 12:07:23 -05:00
GiteaMirror commented 2026-04-12 12:07:24 -05:00
Author
Owner
Copy Link

@charmandercha commented on GitHub (Mar 6, 2025):

any news on this?

<!-- gh-comment-id:2704733659 --> @charmandercha commented on GitHub (Mar 6, 2025): any news on this?
GiteaMirror commented 2026-04-12 12:07:25 -05:00
Author
Owner
Copy Link

@n00mkrad commented on GitHub (Mar 9, 2025):

bump

<!-- gh-comment-id:2709053396 --> @n00mkrad commented on GitHub (Mar 9, 2025): bump
GiteaMirror commented 2026-04-12 12:07:27 -05:00
Author
Owner
Copy Link

@dmarx commented on GitHub (Apr 15, 2025):

bump

<!-- gh-comment-id:2807714862 --> @dmarx commented on GitHub (Apr 15, 2025): bump
GiteaMirror commented 2026-04-12 12:07:28 -05:00
Author
Owner
Copy Link

@FullstackSensei commented on GitHub (Apr 16, 2025):

bump

<!-- gh-comment-id:2807850441 --> @FullstackSensei commented on GitHub (Apr 16, 2025): bump
GiteaMirror commented 2026-04-12 12:07:29 -05:00
Author
Owner
Copy Link

@b8zhong commented on GitHub (Apr 16, 2025):

Bump

<!-- gh-comment-id:2807938935 --> @b8zhong commented on GitHub (Apr 16, 2025): Bump
GiteaMirror commented 2026-04-12 12:07:30 -05:00
Author
Owner
Copy Link

@wviana commented on GitHub (Apr 16, 2025):

Bump

<!-- gh-comment-id:2807970995 --> @wviana commented on GitHub (Apr 16, 2025): Bump
GiteaMirror commented 2026-04-12 12:07:31 -05:00
Author
Owner
Copy Link

@realcoloride commented on GitHub (Apr 16, 2025):

bump

<!-- gh-comment-id:2809983837 --> @realcoloride commented on GitHub (Apr 16, 2025): bump
GiteaMirror commented 2026-04-12 12:07:32 -05:00
Author
Owner
Copy Link

@maxxu123456 commented on GitHub (Apr 22, 2025):

bump

<!-- gh-comment-id:2821490616 --> @maxxu123456 commented on GitHub (Apr 22, 2025): bump
GiteaMirror commented 2026-04-12 12:07:32 -05:00
Author
Owner
Copy Link

@arch-btw commented on GitHub (May 1, 2025):

411 days have passed since this issue was opened, with no response from the ollama maintainers. It's disconcerting to see how little credit is given to llama.cpp, with its sole mention in the README (all the way down at the bottom) as a "supported backend" - a glaring underestimation of the significance of llama.cpp in the software's functionality.

This lack of recognition raises serious concerns about the licensing issues at play. I'm also wondering about the true intentions behind ollama's development. Is the goal to eventually sell or profit from ollama while maintaining a veneer of independence this entire time?

@jmorganca @mchiang0610 @ParthSareen @BruceMacD

<!-- gh-comment-id:2844871491 --> @arch-btw commented on GitHub (May 1, 2025): 411 days have passed since this issue was opened, with no response from the ollama maintainers. It's disconcerting to see how little credit is given to [llama.cpp](https://github.com/ggml-org/llama.cpp), with its sole mention in the README (all the way down at the bottom) as a "supported backend" - a glaring underestimation of the significance of [llama.cpp](https://github.com/ggml-org/llama.cpp) in the software's functionality. This lack of recognition raises serious concerns about the licensing issues at play. I'm also wondering about the true intentions behind ollama's development. Is the goal to eventually sell or profit from ollama while maintaining a veneer of independence this entire time? @jmorganca @mchiang0610 @ParthSareen @BruceMacD
GiteaMirror commented 2026-04-12 12:07:33 -05:00
Author
Owner
Copy Link

@ghost commented on GitHub (May 16, 2025):

bump

<!-- gh-comment-id:2886606657 --> @ghost commented on GitHub (May 16, 2025): bump
GiteaMirror commented 2026-04-12 12:07:34 -05:00
Author
Owner
Copy Link

@prinpal commented on GitHub (May 16, 2025):

bump

<!-- gh-comment-id:2886656912 --> @prinpal commented on GitHub (May 16, 2025): bump
GiteaMirror commented 2026-04-12 12:07:34 -05:00
Author
Owner
Copy Link

@DanielFerguson commented on GitHub (May 16, 2025):

bump

<!-- gh-comment-id:2886669284 --> @DanielFerguson commented on GitHub (May 16, 2025): bump
GiteaMirror commented 2026-04-12 12:07:36 -05:00
Author
Owner
Copy Link

@hynet-mel commented on GitHub (May 16, 2025):

The only reference mentioning Georgi Gerganov I see is in the README.md, might be a good idea to give some credit somewhere else too. But I see atleast in the main README.md there, that's something atleast

<!-- gh-comment-id:2886675740 --> @hynet-mel commented on GitHub (May 16, 2025): The only reference mentioning Georgi Gerganov I see is in the README.md, might be a good idea to give some credit somewhere else too. But I see atleast in the main README.md there, that's something atleast
GiteaMirror commented 2026-04-12 12:07:36 -05:00
Author
Owner
Copy Link

@IvanVnucec commented on GitHub (May 16, 2025):

This is such a stupid issue. Who cares...

<!-- gh-comment-id:2886690712 --> @IvanVnucec commented on GitHub (May 16, 2025): This is such a stupid issue. Who cares...
GiteaMirror commented 2026-04-12 12:07:37 -05:00
Author
Owner
Copy Link

@ivankrut856 commented on GitHub (May 16, 2025):

@jart, llama.cpp rights holders should go ahead and sue already

<!-- gh-comment-id:2886726138 --> @ivankrut856 commented on GitHub (May 16, 2025): @jart, llama.cpp rights holders should go ahead and sue already
GiteaMirror commented 2026-04-12 12:07:38 -05:00
Author
Owner
Copy Link

@mbanzi commented on GitHub (May 16, 2025):

This is such a stupid issue. Who cares...

Actually a lot of us care, if you want to use open source code you need to follow the rules.

<!-- gh-comment-id:2886793399 --> @mbanzi commented on GitHub (May 16, 2025): > This is such a stupid issue. Who cares... Actually a lot of us care, if you want to use open source code you need to follow the rules.
GiteaMirror commented 2026-04-12 12:07:39 -05:00
Author
Owner
Copy Link

@neopunisher commented on GitHub (May 16, 2025):

bump do the right thing

<!-- gh-comment-id:2886793580 --> @neopunisher commented on GitHub (May 16, 2025): bump do the right thing
GiteaMirror commented 2026-04-12 12:07:40 -05:00
Author
Owner
Copy Link

@levifig commented on GitHub (May 16, 2025):

https://github.com/ollama/ollama/blob/main/ml/backend/ggml/ggml/LICENSE

The MIT License in llama.cpp doesn't include Georgi's name but "ggml authors".

<!-- gh-comment-id:2886867013 --> @levifig commented on GitHub (May 16, 2025): https://github.com/ollama/ollama/blob/main/ml/backend/ggml/ggml/LICENSE The MIT License in `llama.cpp` doesn't include Georgi's name but "ggml authors".
GiteaMirror commented 2026-04-12 12:07:40 -05:00
Author
Owner
Copy Link

@erikli commented on GitHub (May 16, 2025):

https://github.com/ollama/ollama/blob/main/ml/backend/ggml/ggml/LICENSE

The MIT License in llama.cpp doesn't include Georgi's name but "ggml authors".

Yes, the singular author name was updated to "ggml authors" in the LICENSE at e11a8999b5 which is dated Apr 9, 2024, which also added a generated AUTHORS file.

<!-- gh-comment-id:2887070546 --> @erikli commented on GitHub (May 16, 2025): > https://github.com/ollama/ollama/blob/main/ml/backend/ggml/ggml/LICENSE > > The MIT License in `llama.cpp` doesn't include Georgi's name but "ggml authors". Yes, the singular author name was updated to "ggml authors" in the LICENSE at https://github.com/ggml-org/llama.cpp/commit/e11a8999b5690f810c2c99c14347f0834e68c524 which is dated Apr 9, 2024, which also added a generated AUTHORS file.
GiteaMirror commented 2026-04-12 12:07:42 -05:00
Author
Owner
Copy Link

@sebastianmuell commented on GitHub (May 16, 2025):

Bump

<!-- gh-comment-id:2887147059 --> @sebastianmuell commented on GitHub (May 16, 2025): Bump
GiteaMirror commented 2026-04-12 12:07:43 -05:00
Author
Owner
Copy Link

@ErfolgreichCharismatisch commented on GitHub (May 16, 2025):

This is such a stupid issue. Who cares...

I am totally with him, who gives a shit. "Mimimi I created it but some people did not mention my name. I mean I am totally for open source and free software, helping people" but in reality those people only seem to have done it for advertising and bragging purposes and now that fell flat. Besides, who even reads that other than some nerds who are all about principles so they can be publicly "outraged" and virtue signal. Go downvote me, you nerds, if you disagree, I don't care.

<!-- gh-comment-id:2887148518 --> @ErfolgreichCharismatisch commented on GitHub (May 16, 2025): > This is such a stupid issue. Who cares... I am totally with him, who gives a shit. "Mimimi I created it but some people did not mention my name. I mean I am totally for open source and free software, helping people" but in reality those people only seem to have done it for advertising and bragging purposes and now that fell flat. Besides, who even reads that other than some nerds who are all about principles so they can be publicly "outraged" and virtue signal. Go downvote me, you nerds, if you disagree, I don't care.
GiteaMirror commented 2026-04-12 12:07:44 -05:00
Author
Owner
Copy Link

@jeffattridge commented on GitHub (May 16, 2025):

This is such a stupid issue. Who cares...

I am totally with him, who gives a shit. "Mimimi I created it but some people did not mention my name. I mean I am totally for open source and free software, helping people" but in reality those people only seem to have done it for advertising and bragging purposes and now that fell flat. Besides, who even reads that other than some nerds who are all about principles so they can be publicly "outraged" and virtue signal. Go downvote me, you nerds, if you disagree, I don't care.

Imagine being this loledgy. Please find somewhere else to troll, I think 4chan is back up.

<!-- gh-comment-id:2887312342 --> @jeffattridge commented on GitHub (May 16, 2025): > > This is such a stupid issue. Who cares... > > I am totally with him, who gives a shit. "Mimimi I created it but some people did not mention my name. I mean I am totally for open source and free software, helping people" but in reality those people only seem to have done it for advertising and bragging purposes and now that fell flat. Besides, who even reads that other than some nerds who are all about principles so they can be publicly "outraged" and virtue signal. Go downvote me, you nerds, if you disagree, I don't care. Imagine being this loledgy. Please find somewhere else to troll, I think 4chan is back up.
GiteaMirror commented 2026-04-12 12:07:44 -05:00
Author
Owner
Copy Link

@danielealbano commented on GitHub (May 16, 2025):

I love people saying that doesn't matter, that it's a stupid issue... So blindfolded...

Do you actually realize that who built llama.cpp did it for fun/money/passion and to be recognized?

Take away the second part and have people appropriate it ILLEGALLY (yes, an OSS license is to define the license under which the code is copyrighted, breaking it means using copyrighted material illegally) and the result is that the author will lose a good share of motivation.

Do you want the authors behind llama.cpp stop working on it? If I would be then, I would have already changed the license to AGPL or similar....

<!-- gh-comment-id:2887332632 --> @danielealbano commented on GitHub (May 16, 2025): I love people saying that doesn't matter, that it's a stupid issue... So blindfolded... Do you actually realize that who built llama.cpp did it for fun/money/passion and to be recognized? Take away the second part and have people appropriate it ILLEGALLY (yes, an OSS license is to define the license under which the code is copyrighted, breaking it means using copyrighted material illegally) and the result is that the author will lose a good share of motivation. Do you want the authors behind llama.cpp stop working on it? If I would be then, I would have already changed the license to AGPL or similar....
GiteaMirror commented 2026-04-12 12:07:45 -05:00
Author
Owner
Copy Link

@rsohlot commented on GitHub (May 16, 2025):

Let's create a new clone and maintain that ,which will follows the MIT rules

<!-- gh-comment-id:2887336403 --> @rsohlot commented on GitHub (May 16, 2025): Let's create a new clone and maintain that ,which will follows the MIT rules
GiteaMirror commented 2026-04-12 12:07:46 -05:00
Author
Owner
Copy Link

@nagesh-c commented on GitHub (May 16, 2025):

bump

<!-- gh-comment-id:2887508874 --> @nagesh-c commented on GitHub (May 16, 2025): bump
GiteaMirror commented 2026-04-12 12:07:47 -05:00
Author
Owner
Copy Link

@daniel-dona commented on GitHub (May 16, 2025):

Is there any PR for this already? If not I guess it could be easy to just include the LICENSE files that are already in the repository.

Releases are built by this workflow https://github.com/ollama/ollama/blob/main/.github/workflows/release.yaml

<!-- gh-comment-id:2887563799 --> @daniel-dona commented on GitHub (May 16, 2025): Is there any PR for this already? If not I guess it could be easy to just include the `LICENSE` files that are already in the repository. Releases are built by this workflow https://github.com/ollama/ollama/blob/main/.github/workflows/release.yaml
GiteaMirror commented 2026-04-12 12:07:48 -05:00
Author
Owner
Copy Link

@reverofevil commented on GitHub (May 16, 2025):

This has a bit more to it than merely "ollama violates license of llama.cpp".

The whole web is built around MIT libraries, dozens of them on every page. What was the last time you've seen a page with a list of MIT licenses of libraries used on a website? Make an experiment: count how many minutes it will take to find Github's page that shows React's MIT license, which it clearly uses to render this page.

Sadly violating terms of MIT license is a status-quo.

<!-- gh-comment-id:2887705559 --> @reverofevil commented on GitHub (May 16, 2025): This has a bit more to it than merely "ollama violates license of llama.cpp". The whole web is built around MIT libraries, dozens of them on every page. What was the last time you've seen a page with a list of MIT licenses of libraries used on a website? Make an experiment: count how many minutes it will take to find Github's page that shows [React's MIT license](https://github.com/facebook/react/blob/main/LICENSE), which it clearly uses to render this page. Sadly violating terms of MIT license is a status-quo.
GiteaMirror commented 2026-04-12 12:07:49 -05:00
Author
Owner
Copy Link

@jart commented on GitHub (May 16, 2025):

When you use MIT/ISC/BSD/etc. code in your website, you should make sure your JS/CSS/HTML minifier doesn't strip away the copyright notice. For example, the Closure Compiler has a @license JSDoc tag for doing just that. You're not required to visibly mention the copyright notice on the web page.

<!-- gh-comment-id:2887734918 --> @jart commented on GitHub (May 16, 2025): When you use MIT/ISC/BSD/etc. code in your website, you should make sure your JS/CSS/HTML minifier doesn't strip away the copyright notice. For example, the Closure Compiler has a `@license` JSDoc tag for doing just that. You're not required to visibly mention the copyright notice on the web page.
GiteaMirror commented 2026-04-12 12:07:50 -05:00
Author
Owner
Copy Link

@3y4m4r1n commented on GitHub (May 16, 2025):

The less stupid copyright and license rubbish I see in my software the better

<!-- gh-comment-id:2887801112 --> @3y4m4r1n commented on GitHub (May 16, 2025): The less stupid copyright and license rubbish I see in my software the better
GiteaMirror commented 2026-04-12 12:07:51 -05:00
Author
Owner
Copy Link

@reverofevil commented on GitHub (May 17, 2025):

you should make sure

Well, we should, but there is no mention of "Meta Platforms" in source code of this page either. Worse, all the GPL software was likely used to train LLMs (and early jailbreaks even proved it), yet we don't see Copilot or Chatgpt being sued.

Licenses only matter if you're ready and capable of enforcing them. Taking on legal matters with non-legal tools just doesn't work. You're a contributor of llama.cpp, so consulting a lawyer and sending a legal letter would be a good start. IANAL, but also Github admits DMCA takedowns.

<!-- gh-comment-id:2888368407 --> @reverofevil commented on GitHub (May 17, 2025): > you should make sure Well, _we should_, but there is no mention of "Meta Platforms" in source code of this page either. Worse, all the GPL software was likely used to train LLMs (and early jailbreaks even proved it), yet we don't see Copilot or Chatgpt being sued. Licenses only matter if you're ready and capable of enforcing them. Taking on legal matters with non-legal tools just doesn't work. You're a contributor of llama.cpp, so consulting a lawyer and sending a legal letter would be a good start. IANAL, but also Github admits DMCA takedowns.
GiteaMirror commented 2026-04-12 12:07:52 -05:00
Author
Owner
Copy Link

@Delta456 commented on GitHub (May 18, 2025):

bump

<!-- gh-comment-id:2888875110 --> @Delta456 commented on GitHub (May 18, 2025): bump
GiteaMirror commented 2026-04-12 12:07:53 -05:00
Author
Owner
Copy Link

@Netherquark commented on GitHub (May 18, 2025):

bump

<!-- gh-comment-id:2889147791 --> @Netherquark commented on GitHub (May 18, 2025): bump
GiteaMirror commented 2026-04-12 12:07:54 -05:00
Author
Owner
Copy Link

@Fr0d0Beutl1n commented on GitHub (May 20, 2025):

@jmorganca @mchiang0610 @ParthSareen @BruceMacD

Are you deliberately ignoring licensing issues?

<!-- gh-comment-id:2893330853 --> @Fr0d0Beutl1n commented on GitHub (May 20, 2025): @jmorganca @mchiang0610 @ParthSareen @BruceMacD Are you deliberately ignoring licensing issues?
GiteaMirror commented 2026-04-12 12:07:54 -05:00
Author
Owner
Copy Link

@ErfolgreichCharismatisch commented on GitHub (May 20, 2025):

This is the first time I unsubscribe from a thread for it being utterly pointless.

<!-- gh-comment-id:2893612442 --> @ErfolgreichCharismatisch commented on GitHub (May 20, 2025): This is the first time I unsubscribe from a thread for it being utterly pointless.
GiteaMirror commented 2026-04-12 12:07:56 -05:00
Author
Owner
Copy Link

@perfectecologietool commented on GitHub (May 20, 2025):

https://github.com/ollama/ollama/blob/main/llama/llama.cpp/LICENSE key isnt authorship recognition but your fellow person's ability to tinker, modify,improve it, and copyright notice persistence allows a feedback loop back to the original author who can see your improvements. Tangibly practical free lunches.

<!-- gh-comment-id:2893880706 --> @perfectecologietool commented on GitHub (May 20, 2025): https://github.com/ollama/ollama/blob/main/llama/llama.cpp/LICENSE key isnt authorship recognition but your fellow person's ability to tinker, modify,improve it, and copyright notice persistence allows a feedback loop back to the original author who can see your improvements. Tangibly practical free lunches.
GiteaMirror commented 2026-04-12 12:07:56 -05:00
Author
Owner
Copy Link

@jart commented on GitHub (Jun 10, 2025):

Founders of Ollama @jmorganca and @mchiang0610.

I brought this license compliance issue to your attention over a year ago and I still haven't heard any response. It appears that news of this issue went viral on several Internet forums a month ago. Many voices have since chimed in lending their support. Yet this issue continues to languish, and I'm beginning to grow impatient.

If you're unfamiliar with me and who I am, I'm the author of your matrix multiplication code. This is the algorithm at the very heart of what your software does. I believe I speak for myself and many others when I say we admire the blockbuster hit you've achieved in AI with Ollama. We're happy you've found our work useful. We're thrilled that it's helped you earn fame and fortune. All we ask in return is to be able to watch your film with our friends and point our finger at screen at the end and say, "See my name up there? I was there. I helped make that!" But unfortunately your film forgot to include the credits scene.

That's the basic idea behind what it means to comply with the MIT license and I want you to make fixing this your top priority.

The first thing I want to see happen is for a flag to be added that prints the license and the full text of the llama.cpp AUTHORS file which lists the copyright holders. I want the change to be authored by one of you two personally, so everyone can see that Ollama's respect for the community comes from the very top. This should take no more than an hour of your time, but it will accomplish the majority of what's needed to mend things. Afterwards, we should all perform a more detailed analysis of Ollama's dependencies to determine what other works require attribution and how best it should be presented.

<!-- gh-comment-id:2957772566 --> @jart commented on GitHub (Jun 10, 2025): Founders of Ollama @jmorganca and @mchiang0610. I brought this license compliance issue to your attention over a year ago and I still haven't heard any response. It appears that news of this issue went viral on several Internet forums a month ago. Many voices have since chimed in lending their support. Yet this issue continues to languish, and I'm beginning to grow impatient. If you're unfamiliar with me and who I am, I'm the author of your matrix multiplication code. This is the algorithm at the very heart of what your software does. I believe I speak for myself and many others when I say we admire the blockbuster hit you've achieved in AI with Ollama. We're happy you've found our work useful. We're thrilled that it's helped you earn fame and fortune. All we ask in return is to be able to watch your film with our friends and point our finger at screen at the end and say, "See my name up there? I was there. I helped make that!" But unfortunately your film forgot to include the credits scene. That's the basic idea behind what it means to comply with the MIT license and I want you to make fixing this your top priority. The first thing I want to see happen is for a flag to be added that prints the license and the full text of the [llama.cpp AUTHORS file](https://github.com/ggml-org/llama.cpp/blob/master/AUTHORS) which lists the copyright holders. I want the change to be authored by one of you two personally, so everyone can see that Ollama's respect for the community comes from the very top. This should take no more than an hour of your time, but it will accomplish the majority of what's needed to mend things. Afterwards, we should all perform a more detailed analysis of Ollama's dependencies to determine what other works require attribution and how best it should be presented.
GiteaMirror commented 2026-04-12 12:07:57 -05:00
Author
Owner
Copy Link

@SamuelTallet commented on GitHub (Jun 11, 2025):

Please rename this project to ollama.cpp for clear attribution 😄
No seriously, including third-party LICENSE files cost nothing...

<!-- gh-comment-id:2963384396 --> @SamuelTallet commented on GitHub (Jun 11, 2025): Please rename this project to _ollama.cpp_ for clear attribution 😄 No seriously, including third-party LICENSE files cost nothing...
GiteaMirror commented 2026-04-12 12:07:58 -05:00
Author
Owner
Copy Link

@corvec commented on GitHub (Jun 11, 2025):

EDIT 2: llama.cpp does bundle its own LICENSE file, so this is a non-point.

As llama.cpp does not bundle its own LICENSE, does that mean that every person, other than the copyright holders, who downloads or otherwise copies it; who forks and builds it (without changing any code); forks and modifies (without bundling the LICENSE notice themselves); or who publishes or distributes a release of it, is also out of compliance with the llama.cpp license?

EDIT: Fixed the second "fork" operation that would put you out of compliance.

<!-- gh-comment-id:2964504140 --> @corvec commented on GitHub (Jun 11, 2025): EDIT 2: llama.cpp does bundle its own LICENSE file, so this is a non-point. As llama.cpp does not bundle its own `LICENSE`, does that mean that every person, other than the copyright holders, who downloads or otherwise copies it; who forks and builds it (without changing any code); forks and modifies (without bundling the LICENSE notice themselves); or who publishes or distributes a release of it, is also out of compliance with the llama.cpp license? EDIT: Fixed the second "fork" operation that would put you out of compliance.
GiteaMirror commented 2026-04-12 12:07:58 -05:00
Author
Owner
Copy Link

@beaugunderson commented on GitHub (Jun 11, 2025):

@corvec not sure what you mean, llama.cpp has a LICENSE file right here

<!-- gh-comment-id:2964513488 --> @beaugunderson commented on GitHub (Jun 11, 2025): @corvec not sure what you mean, llama.cpp [has a LICENSE file right here](https://github.com/ggml-org/llama.cpp/blob/master/LICENSE)
GiteaMirror commented 2026-04-12 12:07:59 -05:00
Author
Owner
Copy Link

@corvec commented on GitHub (Jun 11, 2025):

@corvec not sure what you mean, llama.cpp has a LICENSE file right here

If that were sufficient then Ollama would be fine due to reproducing that here.

This is what I was referencing from the MIT license (emphasis mine):

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

That said, I downloaded one of the most recent Llama.cpp releases and I was wrong - the archive file includes the license text (though not the authors file).

<!-- gh-comment-id:2964541262 --> @corvec commented on GitHub (Jun 11, 2025): > [@corvec](https://github.com/corvec) not sure what you mean, llama.cpp [has a LICENSE file right here](https://github.com/ggml-org/llama.cpp/blob/master/LICENSE) If that were sufficient then Ollama would be fine due to reproducing that [here](https://github.com/ollama/ollama/blob/main/llama/llama.cpp/LICENSE). This is what I was referencing from the MIT license (emphasis mine): > The above copyright notice and this permission notice shall be included in **all** > **copies or substantial portions** of the Software. That said, I downloaded one of the most recent Llama.cpp releases and I was wrong - the archive file includes the license text (though not the authors file).
GiteaMirror commented 2026-04-12 12:08:00 -05:00
Author
Owner
Copy Link

@JerryI commented on GitHub (Jul 8, 2025):

No seriously, including third-party LICENSE files cost nothing...

No, it seriously costs tons of ego for ollama devs. Prove me wrong.

PS: Most arguments seem like kindergarten level

<!-- gh-comment-id:3050131914 --> @JerryI commented on GitHub (Jul 8, 2025): > No seriously, including third-party LICENSE files cost nothing... No, it seriously costs tons of ego for ollama devs. Prove me wrong. PS: Most arguments seem like kindergarten level
GiteaMirror commented 2026-04-12 12:08:02 -05:00
Author
Owner
Copy Link

@sbera77 commented on GitHub (Aug 11, 2025):

Bump

<!-- gh-comment-id:3175620514 --> @sbera77 commented on GitHub (Aug 11, 2025): Bump
GiteaMirror commented 2026-04-12 12:08:04 -05:00
Author
Owner
Copy Link

@ductai199x commented on GitHub (Aug 11, 2025):

Bump!

<!-- gh-comment-id:3176686970 --> @ductai199x commented on GitHub (Aug 11, 2025): Bump!
GiteaMirror commented 2026-04-12 12:08:05 -05:00
Author
Owner
Copy Link

@n00mkrad commented on GitHub (Aug 12, 2025):

Radio silence means they know they're guilty. Bump.

<!-- gh-comment-id:3177883562 --> @n00mkrad commented on GitHub (Aug 12, 2025): Radio silence means they know they're guilty. Bump.
GiteaMirror commented 2026-04-12 12:08:06 -05:00
Author
Owner
Copy Link

@3eeps commented on GitHub (Aug 12, 2025):

pretty sad, isn't it?

<!-- gh-comment-id:3178098732 --> @3eeps commented on GitHub (Aug 12, 2025): pretty sad, isn't it?
GiteaMirror commented 2026-04-12 12:08:07 -05:00
Author
Owner
Copy Link

@Exotik850 commented on GitHub (Aug 12, 2025):

It’s honestly surprising-and disappointing-that this is still unresolved.

Forgetting to link to a license or missing some attribution can happen by accident; those are usually quick, good-faith mistakes. But going over a year without any clear acknowledgment, timeline, or plan to address it sends a very different message. At that point, it’s not just an oversight; it starts to look like a conscious decision to ignore a core principle of open source: respecting and crediting the work you build on.

Open source only works because of mutual trust and reciprocity. When a project breaks that trust, it puts the whole ecosystem at risk. For me, continuing to use or support this without change would feel like endorsing that behavior, and that’s a line I can’t cross.

Switching to llama.cpp and llama-swap immediately.
If a project can’t honor the license it’s built on, can it truly call itself open source?

<!-- gh-comment-id:3179754397 --> @Exotik850 commented on GitHub (Aug 12, 2025): It’s honestly surprising-and disappointing-that this is still unresolved. Forgetting to link to a license or missing some attribution can happen by accident; those are usually quick, good-faith mistakes. But going [over a year](https://github.com/ollama/ollama/issues/3185#issue-2190192859) without any clear acknowledgment, timeline, or plan to address it sends a very different message. At that point, it’s not just an oversight; it starts to look like a conscious decision to ignore a core principle of open source: _respecting and crediting the work you build on_. Open source only works because of mutual trust and reciprocity. When a project breaks that trust, it puts the whole ecosystem at risk. For me, continuing to use or support this without change would feel like endorsing that behavior, and that’s a line I can’t cross. Switching to llama.cpp and llama-swap immediately. If a project can’t honor the license it’s built on, can it truly call itself open source?
GiteaMirror commented 2026-04-12 12:08:08 -05:00
Author
Owner
Copy Link

@he29-net commented on GitHub (Aug 17, 2025):

Given how trivial it would be to fix this issue and how long ago it was opened, at this point I can only assume that the Ollama project violates the license terms willfully and intentionally.

That is not cool, and not a behavior that helps to build a thriving open source ecosystem, which is what made Ollama possible in the first place.

Until this is fixed, I would suggest supporting and recommending alternatives of Ollama instead.

<!-- gh-comment-id:3194593989 --> @he29-net commented on GitHub (Aug 17, 2025): Given how trivial it would be to fix this issue and how long ago it was opened, at this point I can only assume that the Ollama project violates the license terms willfully and intentionally. That is not cool, and not a behavior that helps to build a thriving open source ecosystem, which is what made Ollama possible in the first place. Until this is fixed, I would suggest supporting and recommending alternatives of Ollama instead.
GiteaMirror commented 2026-04-12 12:08:08 -05:00
Author
Owner
Copy Link

@SuperUserNameMan commented on GitHub (Aug 18, 2025):

Naive questions :

  1. since the llama.cpp license contains no names (it contains "(c) The ggml authors"), does it mean that any contributor (including those who pushed a minor PR to correct some misspellings and typos) are considered "ggml authors" ?

  2. as I pushed a minor PR in llama.cpp (a simple llama.cpp example), does it make me part of the "ggml authors" ?

  3. since ollama already provides a copy of the llama.cpp license into its source code, isn't it okay already ?

  4. and since only users who really want to know the recipe care about the ingredients, isn't the ollama source-code repo the best location to put this license ?

<!-- gh-comment-id:3195758872 --> @SuperUserNameMan commented on GitHub (Aug 18, 2025): Naive questions : 1) since the llama.cpp license contains no names (it contains "(c) The ggml authors"), does it mean that any contributor (including those who pushed a minor PR to correct some misspellings and typos) are considered "ggml authors" ? 2) as I pushed a minor PR in llama.cpp (a simple llama.cpp example), does it make me part of the "ggml authors" ? 3) since ollama already provides a copy of the llama.cpp license into its source code, isn't it okay already ? 4) and since only users who really want to know the recipe care about the ingredients, isn't the ollama source-code repo the best location to put this license ?
GiteaMirror commented 2026-04-12 12:08:09 -05:00
Author
Owner
Copy Link

@Exotik850 commented on GitHub (Aug 18, 2025):

To answer @SuperUserNameMan’s questions:

  1. Since the llama.cpp license contains no names (it just says “(c) The ggml authors”), does this mean that any contributor (even for minor typo fixes) is considered a “ggml author”?

Yes. By contributing to an open-source project, you become one of its “authors,” regardless of how large or small your contribution is. The code you write remains yours, but once merged it’s also covered by the project’s license. See GitHub’s Terms of Service for details.

  1. As I pushed a small PR to llama.cpp, does that make me part of the “ggml authors”?

Yes, see above.

  1. Since ollama already provides a copy of the llama.cpp license in its source code, isn’t that enough?

Not quite. The MIT license requires that the license be included with all copies or substantial portions of the software. That includes binaries, not just the source repo. While it’s often overlooked in practice, leaving it out is technically a license violation. It’s usually an easy oversight to fix.

  1. Since only users who care about the “recipe” care about the license, isn’t the source repo the best place to put it?

Again, see above; the license must travel with every distribution, not only the source repo.

<!-- gh-comment-id:3197190750 --> @Exotik850 commented on GitHub (Aug 18, 2025): To answer @SuperUserNameMan’s questions: > 1. Since the llama.cpp license contains no names (it just says “(c) The ggml authors”), does this mean that any contributor (even for minor typo fixes) is considered a “ggml author”? Yes. By contributing to an open-source project, you become one of its “authors,” regardless of how large or small your contribution is. The code you write remains yours, but once merged it’s also covered by the project’s license. See [GitHub’s Terms of Service](https://docs.github.com/en/site-policy/github-terms/github-terms-of-service#6-contributions-under-repository-license) for details. > 2. As I pushed a small PR to llama.cpp, does that make me part of the “ggml authors”? Yes, see above. > 3. Since ollama already provides a copy of the llama.cpp license in its source code, isn’t that enough? Not quite. The MIT license requires that the license be included with *all copies or substantial portions* of the software. That includes binaries, not just the source repo. While it’s often overlooked in practice, leaving it out is technically a license violation. It’s usually an easy oversight to fix. > 4. Since only users who care about the “recipe” care about the license, isn’t the source repo the best place to put it? Again, see above; the license must travel with every distribution, not only the source repo.
GiteaMirror commented 2026-04-12 12:08:10 -05:00
Author
Owner
Copy Link

@SuperUserNameMan commented on GitHub (Aug 18, 2025):

@Exotik850 : thank you for your answers.

<!-- gh-comment-id:3197370874 --> @SuperUserNameMan commented on GitHub (Aug 18, 2025): @Exotik850 : thank you for your answers.
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 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#1963
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?