github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 17:49:24 -05:00
Code Issues 14.6k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #2864] Read and write files (To allow for more types of AI models in Ollama) #1748

New Issue
Closed
opened 2026-04-12 11:44:08 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 11:44:08 -05:00
Owner
Copy Link

Originally created by @trymeouteh on GitHub (Mar 1, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2864

To my understanding it will seem Ollama can only support text chat based models and to my understand it could support more kinds of models by adding a reading files and writing files feature...

  • A way to have a model read one or more file on your system and be able to send it to the AI model to process.
  • A way for the AI model to return a file, by generating a file (Writing a file) onto the system and them being able to link it into the chat response.

To my understanding, this can allow Ollama to work with more model types such as text to image models, image to image models, text to audio models, audio to text models, audio to audio models, etc.

Originally created by @trymeouteh on GitHub (Mar 1, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2864 To my understanding it will seem Ollama can only support text chat based models and to my understand it could support more kinds of models by adding a reading files and writing files feature... - A way to have a model read one or more file on your system and be able to send it to the AI model to process. - A way for the AI model to return a file, by generating a file (Writing a file) onto the system and them being able to link it into the chat response. To my understanding, this can allow Ollama to work with more model types such as text to image models, image to image models, text to audio models, audio to text models, audio to audio models, etc.
GiteaMirror commented 2026-04-12 11:44:09 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 1, 2024):

For passing a file, you can do that in the REPL by passing the filename which the client will read and then pass to the API. The API already supports this (in both the /api/chat and /api/generate` endpoints). We don't yet have a way for the LLM to return a file, but we'd have to support it down the road when we support other models like stable diffusion.

<!-- gh-comment-id:1973630379 --> @pdevine commented on GitHub (Mar 1, 2024): For passing a file, you can do that in the REPL by passing the filename which the client will read and then pass to the API. The API already supports this (in both the `/api/chat` and /api/generate` endpoints). We don't *yet* have a way for the LLM to return a file, but we'd have to support it down the road when we support other models like stable diffusion.
GiteaMirror commented 2026-04-12 11:44:10 -05:00
Author
Owner
Copy Link

@trymeouteh commented on GitHub (Mar 7, 2024):

Is it already possible to have a model read two or more files or just one file at a time in a chat request?

<!-- gh-comment-id:1984149870 --> @trymeouteh commented on GitHub (Mar 7, 2024): Is it already possible to have a model read two or more files or just one file at a time in a chat request?
GiteaMirror commented 2026-04-12 11:44:10 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 7, 2024):

I'm realizing on rereading my response that I was not super clear. The API doesn't read the file per se, but you can pass images along in the request. In the API you can do that with:

curl http://localhost:11434/api/generate -d '{
  "model": "llama2",
  "prompt": "Why is the sky blue?",
  "images": [ ... <list of image data> ... ]
}'

Or you can do:

curl http://localhost:11434/api/chat -d '{
  "model": "llama2",
  "messages": [
    {
      "role": "user",
      "content": "why is the sky blue?",
      "images": [ ... <list of image data> ],
    }
  ]
}'

You can find more info in the docs.

I'm going to go ahead and close the issue, but feel free to keep commenting, or ask away in the discord.

<!-- gh-comment-id:1984217672 --> @pdevine commented on GitHub (Mar 7, 2024): I'm realizing on rereading my response that I was not super clear. The API doesn't read the file per se, but you can pass images along in the request. In the API you can do that with: ``` curl http://localhost:11434/api/generate -d '{ "model": "llama2", "prompt": "Why is the sky blue?", "images": [ ... <list of image data> ... ] }' ``` Or you can do: ``` curl http://localhost:11434/api/chat -d '{ "model": "llama2", "messages": [ { "role": "user", "content": "why is the sky blue?", "images": [ ... <list of image data> ], } ] }' ``` You can find more info in the [docs](https://github.com/ollama/ollama/blob/main/docs/api.md). I'm going to go ahead and close the issue, but feel free to keep commenting, or ask away in the discord.
GiteaMirror commented 2026-04-12 11:44:11 -05:00
Author
Owner
Copy Link

@trymeouteh commented on GitHub (Mar 8, 2024):

Is it possible to pass two or more files into any of the available models for Ollama? If so you can please provide an example without using curl but using the model chat?

<!-- gh-comment-id:1985762420 --> @trymeouteh commented on GitHub (Mar 8, 2024): Is it possible to pass two or more files into any of the available models for Ollama? If so you can please provide an example without using curl but using the model chat?
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#1748
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?