Update README.md to include acknowledgements to llama.cpp #3700
Conversation
|
@survirtual thank you for submitting this. We definitely need to do a better job acknowledging the work of @ggerganov and the llama.cpp team. Maybe we should create a backend engine section? llama.cpp definitely will not be the only engine Ollama uses. We spend a large chunk of time fixing and patching it up to ensure a smooth experience for Ollama users and developers. We don't promise exact features as llama.cpp. This is due to potential breaking changes in it. As a developer tool, we try very hard to keep Ollama running. We don't always get it right, but we try. Overtime, we will be transitioning to more systematically built engines. |
|
added it here! Thank you so much for this PR |
A well documented backend engine section would be a good idea. While there is still only 1 supported engine, I'd really recommend providing front & center llama.cpp acknowledgements (along the vein of the pull request I provided) to curate good will with people who may have a sour taste (along with all the other reasons listed before on the original issue). Once more engines are added, it will be more natural to shift the verbiage around. Many people rely on this project now and it has done wonders to make open-weight LLMs more ergonomic, so I don't like seeing such a simple thing weigh it down. As it stands with the change you've merged, while an improvement, may be perceived badly since it is all the way at the bottom with a single line. It might be better to remove any chance of that negative perception. |
resolves #3697