github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #1721] How to enable Ollama read contents of directory? #63013

New Issue
Closed
opened 2026-05-03 11:15:50 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-03 11:15:50 -05:00
Owner
Copy Link

Originally created by @oliverbob on GitHub (Dec 26, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1721

Since it can read '/home/user/whateverfile.it.is', would it be possible for ollama to be able to read an entire directory or 'repo' for that matter so we can talk to it?

If it is not yet a feature, maybe its neat to add this for developers to help us quickly solve our coding challenges.

Thanks.

Originally created by @oliverbob on GitHub (Dec 26, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1721 Since it can read '/home/user/whateverfile.it.is', would it be possible for ollama to be able to read an entire directory or 'repo' for that matter so we can talk to it? If it is not yet a feature, maybe its neat to add this for developers to help us quickly solve our coding challenges. Thanks.
GiteaMirror commented 2026-05-03 11:15:53 -05:00
Author
Owner
Copy Link

@technovangelist commented on GitHub (Dec 27, 2023):

Hi @oliverbob , thanks for submitting this issue. To read files in to a prompt, you have a few options. First, you can use the features of your shell to pipe in the contents of a file. To read in more than a single file, you need to do a few extra steps because the contents of your files is probably bigger than the context size of the model. So you can use a technique known as RAG. We have a few examples here in our repo that show you how to do RAG with Ollama. Essentially, it comes down to importing your content into some sort of data store, usually in a special format that is semantically searchable. Then you filter the content based on a query. Then that is fed to the model with the prompt and the model generates an answer. The third option is to let someone else build RAG for your. On the front Readme of this repo is a list of community projects. Some of those do various forms of RAG on your files.

Let me know if that answers your questions.

<!-- gh-comment-id:1869970793 --> @technovangelist commented on GitHub (Dec 27, 2023): Hi @oliverbob , thanks for submitting this issue. To read files in to a prompt, you have a few options. First, you can use the features of your shell to pipe in the contents of a file. To read in more than a single file, you need to do a few extra steps because the contents of your files is probably bigger than the context size of the model. So you can use a technique known as RAG. We have a few examples here in our repo that show you how to do RAG with Ollama. Essentially, it comes down to importing your content into some sort of data store, usually in a special format that is semantically searchable. Then you filter the content based on a query. Then that is fed to the model with the prompt and the model generates an answer. The third option is to let someone else build RAG for your. On the front Readme of this repo is a list of community projects. Some of those do various forms of RAG on your files. Let me know if that answers your questions.
GiteaMirror commented 2026-05-03 11:15:53 -05:00
Author
Owner
Copy Link

@xiaoyuvax commented on GitHub (Feb 27, 2024):

Can u be more specific about those third-party tools or sevices? I want Ollama together with any of the models to respond relevantly according to my local documents (maybe extracted by RAG), what exactly should i do to use the RAG?
Ollama cannot access internet or a knowledge base stored in a datebase limits its usability, any way for Ollama to access ElasticSearch or any database for RAG?

<!-- gh-comment-id:1966057952 --> @xiaoyuvax commented on GitHub (Feb 27, 2024): Can u be more specific about those third-party tools or sevices? I want Ollama together with any of the models to respond relevantly according to my local documents (maybe extracted by RAG), what exactly should i do to use the RAG? Ollama cannot access internet or a knowledge base stored in a datebase limits its usability, any way for Ollama to access ElasticSearch or any database for RAG?
GiteaMirror commented 2026-05-03 11:15:55 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (May 10, 2024):

Hi @oliverbob this isn't something Ollama supports today. There are some open issues for having built-in vector store sort of functionality so I'll merge it with that #834

<!-- gh-comment-id:2103639329 --> @jmorganca commented on GitHub (May 10, 2024): Hi @oliverbob this isn't something Ollama supports today. There are some open issues for having built-in vector store sort of functionality so I'll merge it with that #834
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
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#63013
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?