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 #1327] Modelfile prompt should support chat / multiturn. #689

New Issue
Closed
opened 2026-04-12 10:22:07 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 10:22:07 -05:00
Owner
Copy Link

Originally created by @ehartford on GitHub (Nov 30, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1327

Originally assigned to: @BruceMacD on GitHub.

https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#template

So basically all that's coming in is .Prompt which is just a string.

But that can't handle chat and multi turn.

What's coming in should look a messages array. then this template should format that into a prompt.

[
  { "role": "system", "content": "You are a helpful AI assistant" },
  { "role": "user", "content": "Hello AI, How are you today?" },
  { "role": "assistant", "content": "I have no notion of time.  State your question?" },
  { "role": "user", "content": "Oh ok then, tell me the 38th state" }
]

then the template in the modelfile would look something like

{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}<|im_start|>assistant

Basically the idea that a prompt consists of a single system message and a single user message, isn't how most models actually work.

Originally created by @ehartford on GitHub (Nov 30, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1327 Originally assigned to: @BruceMacD on GitHub. https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#template So basically all that's coming in is .Prompt which is just a string. But that can't handle chat and multi turn. What's coming in should look a messages array. then this template should format that into a prompt. ``` [ { "role": "system", "content": "You are a helpful AI assistant" }, { "role": "user", "content": "Hello AI, How are you today?" }, { "role": "assistant", "content": "I have no notion of time. State your question?" }, { "role": "user", "content": "Oh ok then, tell me the 38th state" } ] ``` then the template in the modelfile would look something like ``` {% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}<|im_start|>assistant ``` Basically the idea that a prompt consists of a single system message and a single user message, isn't how most models actually work.
GiteaMirror added the feature request label 2026-04-12 10:22:07 -05:00
GiteaMirror commented 2026-04-12 10:22:07 -05:00
Author
Owner
Copy Link

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

Hey Eric, I just merged a /chat API that will be in the next release which should cover this.

Here's the doc:
https://github.com/jmorganca/ollama/blob/main/docs/api.md#send-chat-messages

It uses messages in the same format that you've suggested here. The template in the Modelfile is still the same right now, but the behavior matches what you suggested here too (templating for each set of messages).

Thanks for the feedback let us know if you notice anything else that doesn't seem right.

<!-- gh-comment-id:1839711562 --> @BruceMacD commented on GitHub (Dec 4, 2023): Hey Eric, I just merged a `/chat` API that will be in the next release which should cover this. Here's the doc: https://github.com/jmorganca/ollama/blob/main/docs/api.md#send-chat-messages It uses messages in the same format that you've suggested here. The template in the Modelfile is still the same right now, but the behavior matches what you suggested here too (templating for each set of messages). Thanks for the feedback let us know if you notice anything else that doesn't seem right.
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#689
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?