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 #215] Function calling #88

New Issue
Closed
opened 2026-04-12 09:37:51 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 09:37:51 -05:00
Owner
Copy Link

Originally created by @nathanleclaire on GitHub (Jul 25, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/215

Trying to get structured/consistent responses out of LLMs can be pretty brutal

OpenAI recently rolled out Function Calling to get the models to stick to pre-defined schemas

it would be excellent if you could specify something like this (ins/outs) in modelfile

FROM llama

INPUT sentence string

ENUM Sentiment ["good", "bad", "neutral"]

OUTPUT classification Sentiment

PROMPT """
You are skilled at detecting tone in user comments.
Classify the following comment:

${sentence}
"""

then something like:

$ ollama run sentiment "ClosedAI has no moat"
bad

(or better yet, with API :) )

Originally created by @nathanleclaire on GitHub (Jul 25, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/215 Trying to get structured/consistent responses out of LLMs can be pretty brutal OpenAI recently rolled out [Function Calling](https://openai.com/blog/function-calling-and-other-api-updates) to get the models to stick to pre-defined schemas it would be excellent if you could specify something like this (ins/outs) in modelfile ``` FROM llama INPUT sentence string ENUM Sentiment ["good", "bad", "neutral"] OUTPUT classification Sentiment PROMPT """ You are skilled at detecting tone in user comments. Classify the following comment: ${sentence} """ ``` then something like: ``` $ ollama run sentiment "ClosedAI has no moat" bad ``` (or better yet, with API :) )
GiteaMirror added the feature request label 2026-04-12 09:37:51 -05:00
GiteaMirror commented 2026-04-12 09:37:54 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Jul 27, 2023):

@nathanleclaire nice! Have you seen this? https://github.com/ggerganov/llama.cpp/pull/1773

<!-- gh-comment-id:1653785533 --> @jmorganca commented on GitHub (Jul 27, 2023): @nathanleclaire nice! Have you seen this? https://github.com/ggerganov/llama.cpp/pull/1773
GiteaMirror commented 2026-04-12 09:37:54 -05:00
Author
Owner
Copy Link

@nathanleclaire commented on GitHub (Jul 27, 2023):

@nathanleclaire nice! Have you seen this? ggerganov/llama.cpp#1773

I had not! That's dope!!

<!-- gh-comment-id:1654549609 --> @nathanleclaire commented on GitHub (Jul 27, 2023): > @nathanleclaire nice! Have you seen this? [ggerganov/llama.cpp#1773](https://github.com/ggerganov/llama.cpp/pull/1773) I had not! That's dope!!
GiteaMirror commented 2026-04-12 09:37:55 -05:00
Author
Owner
Copy Link

@technovangelist commented on GitHub (Dec 4, 2023):

A few weeks back we added format: json which solves most of the points here. We can specify that it should be output as json and we can specify the schema and types to be used. Its not at the modelfile level but can be applied to any model, either through the API or at the CLI with ollama run --format json or in the repl with set format json.

I think this solves your original request. As such I will close the issue. If you think its not solved, let us know what else is needed by reopening the issue. Thanks

<!-- gh-comment-id:1839270628 --> @technovangelist commented on GitHub (Dec 4, 2023): A few weeks back we added `format: json` which solves most of the points here. We can specify that it should be output as json and we can specify the schema and types to be used. Its not at the modelfile level but can be applied to any model, either through the API or at the CLI with `ollama run --format json` or in the repl with `set format json`. I think this solves your original request. As such I will close the issue. If you think its not solved, let us know what else is needed by reopening the issue. Thanks
GiteaMirror commented 2026-04-12 09:37:55 -05:00
Author
Owner
Copy Link

@nathanleclaire commented on GitHub (Dec 7, 2023):

dooooope!

<!-- gh-comment-id:1845519496 --> @nathanleclaire commented on GitHub (Dec 7, 2023): dooooope!
GiteaMirror commented 2026-04-12 09:37:55 -05:00
Author
Owner
Copy Link

@technovangelist commented on GitHub (Dec 9, 2023):

YES!!! That’s the response I hoped for.

<!-- gh-comment-id:1848087071 --> @technovangelist commented on GitHub (Dec 9, 2023): YES!!! That’s the response I hoped for.
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 feature request
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#88
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?