github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #1754] How to add custom LLM models from Huggingface #1005

New Issue
Closed
opened 2026-04-12 10:42:56 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 10:42:56 -05:00
Owner
Copy Link

Originally created by @yiouyou on GitHub (Jan 1, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/1754

I have some fine-tuned models saved on Huggingface. How to add or convert any custome LLM to ollama fitted version?

Originally created by @yiouyou on GitHub (Jan 1, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/1754 I have some fine-tuned models saved on Huggingface. How to add or convert any custome LLM to ollama fitted version?
GiteaMirror added the question label 2026-04-12 10:42:56 -05:00
GiteaMirror commented 2026-04-12 10:42:58 -05:00
Author
Owner
Copy Link

@mongolu commented on GitHub (Jan 2, 2024):

You need to convert it.
Assuming you have all the "requirements" installed and hf model on your local drive, you could inspire from this:
converting hf to gguf
And after successfully converting to gguf, you need to import it in ollama from modelfile

I can confirm I've done it and it works.

<!-- gh-comment-id:1873725209 --> @mongolu commented on GitHub (Jan 2, 2024): You need to convert it. Assuming you have all the "requirements" installed and hf model on your local drive, you could inspire from this: [converting hf to gguf](https://github.com/ggerganov/llama.cpp/blob/edd1ab7bc34c10a780ee7f9a4499f7689cdad36d/scripts/convert-gg.sh#L7) [And after successfully converting to gguf, you need to import it in ollama from modelfile](https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md) I can confirm I've done it and it works.
GiteaMirror commented 2026-04-12 10:42:58 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Jan 2, 2024):

Thanks for opening the issue. Marking this as resolved for now as the question seems to have been answered (thanks to mongolu for that). Please feel free to ask any follow up questions though, I'll keep an eye out.

<!-- gh-comment-id:1873909500 --> @BruceMacD commented on GitHub (Jan 2, 2024): Thanks for opening the issue. Marking this as resolved for now as the question seems to have been answered (thanks to mongolu for that). Please feel free to ask any follow up questions though, I'll keep an eye out.
GiteaMirror commented 2026-04-12 10:42:59 -05:00
Author
Owner
Copy Link

@easp commented on GitHub (Jan 2, 2024):

@mongolu There is a better reference for importing models from HF into Ollama: https://github.com/jmorganca/ollama/blob/main/docs/import.md

This includes instructions on an ollama-provided docker image that makes converting and quantizing a single command.

<!-- gh-comment-id:1874607736 --> @easp commented on GitHub (Jan 2, 2024): @mongolu There is a better reference for importing models from HF into Ollama: https://github.com/jmorganca/ollama/blob/main/docs/import.md This includes instructions on an ollama-provided docker image that makes converting and quantizing a single command.
GiteaMirror commented 2026-04-12 10:43:00 -05:00
Author
Owner
Copy Link

@mongolu commented on GitHub (Jan 2, 2024):

Didn't "find" it (named "import") into "./docs"... :(, so sorry.
Also it's a bit lenghty, but detailed, i presume it's better.
I used a docker image where I had already (or installed) all the necessary deps for this.

<!-- gh-comment-id:1874651962 --> @mongolu commented on GitHub (Jan 2, 2024): Didn't "find" it (named "import") into "./docs"... :(, so sorry. Also it's a bit lenghty, but detailed, i presume it's better. I used a docker image where I had already (or installed) all the necessary deps for this.
GiteaMirror commented 2026-04-12 10:43:01 -05:00
Author
Owner
Copy Link

@ozbillwang commented on GitHub (Jan 28, 2025):

You need to convert it. Assuming you have all the "requirements" installed and hf model on your local drive, you could inspire from this: converting hf to gguf And after successfully converting to gguf, you need to import it in ollama from modelfile

I can confirm I've done it and it works.

this document is not for new model. with current design, we need adjust several files to adapt new model in llama.cpp, I raised a new request in https://github.com/ggerganov/llama.cpp/issues/11460

<!-- gh-comment-id:2617784356 --> @ozbillwang commented on GitHub (Jan 28, 2025): > You need to convert it. Assuming you have all the "requirements" installed and hf model on your local drive, you could inspire from this: [converting hf to gguf](https://github.com/ggerganov/llama.cpp/blob/edd1ab7bc34c10a780ee7f9a4499f7689cdad36d/scripts/convert-gg.sh#L7) [And after successfully converting to gguf, you need to import it in ollama from modelfile](https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md) > > I can confirm I've done it and it works. this document is not for new model. with current design, we need adjust several files to adapt new model in llama.cpp, I raised a new request in https://github.com/ggerganov/llama.cpp/issues/11460
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 question
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#1005
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?