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 #13804] Request for Modelfile for FunctionGemma #34802

New Issue
Closed
opened 2026-04-22 18:40:25 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-22 18:40:25 -05:00
Owner
Copy Link

Originally created by @derrick56007 on GitHub (Jan 20, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/13804

Hi all, I'm wondering if anyone has a working Modelfile for functiongemma.

After following the finetune instructions here, I'm required to use a Modelfile to add the model to ollama.

I've attempting variations of the below but am unable to get tool calling generation to work.

FROM functiongemma-finetune

SYSTEM """You are a model that can do function calling."""

# 2. TEMPLATE: This matches the "developer" role and <escape> logic exactly.
TEMPLATE """{{- if .System }}<start_of_turn>developer
{{ .System }} with the following functions
{{- if .Tools }}
{{- range .Tools }}<start_function_declaration>declaration:{{ .Function.Name }}{description:<escape>{{ .Function.Description }}<escape>,parameters:{{ .Function.Parameters }}}<end_function_declaration>
{{- end }}
{{- end }}<end_of_turn>
{{- end }}
{{- range .Messages }}
{{- if eq .Role "user" }}<start_of_turn>user
{{ .Content }}<end_of_turn>
{{- else if eq .Role "assistant" }}<start_of_turn>model
{{- if .Content }}{{ .Content }}{{ end }}
{{- if .ToolCalls }}
<start_function_call>
{{- range .ToolCalls }}
call:{{ .Function.Name }}{arguments:{{ .Function.Arguments }}}
{{- end }}
<end_function_call>
{{- end }}<end_of_turn>
{{- else if eq .Role "tool" }}<start_of_turn>user
<start_function_response>
response:{{ .Function.Name }}{value:<escape>{{ .Content }}<escape>}
<end_function_response><end_of_turn>
{{- end }}
{{- end }}<start_of_turn>model
"""

PARAMETER temperature 0.0
PARAMETER stop "<end_of_turn>"

However this produces outputs without tool calls

Originally created by @derrick56007 on GitHub (Jan 20, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/13804 Hi all, I'm wondering if anyone has a working Modelfile for functiongemma. After following the finetune instructions [here](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/FunctionGemma_(270M).ipynb), I'm required to use a Modelfile to add the model to ollama. I've attempting variations of the below but am unable to get tool calling generation to work. ``` FROM functiongemma-finetune SYSTEM """You are a model that can do function calling.""" # 2. TEMPLATE: This matches the "developer" role and <escape> logic exactly. TEMPLATE """{{- if .System }}<start_of_turn>developer {{ .System }} with the following functions {{- if .Tools }} {{- range .Tools }}<start_function_declaration>declaration:{{ .Function.Name }}{description:<escape>{{ .Function.Description }}<escape>,parameters:{{ .Function.Parameters }}}<end_function_declaration> {{- end }} {{- end }}<end_of_turn> {{- end }} {{- range .Messages }} {{- if eq .Role "user" }}<start_of_turn>user {{ .Content }}<end_of_turn> {{- else if eq .Role "assistant" }}<start_of_turn>model {{- if .Content }}{{ .Content }}{{ end }} {{- if .ToolCalls }} <start_function_call> {{- range .ToolCalls }} call:{{ .Function.Name }}{arguments:{{ .Function.Arguments }}} {{- end }} <end_function_call> {{- end }}<end_of_turn> {{- else if eq .Role "tool" }}<start_of_turn>user <start_function_response> response:{{ .Function.Name }}{value:<escape>{{ .Content }}<escape>} <end_function_response><end_of_turn> {{- end }} {{- end }}<start_of_turn>model """ PARAMETER temperature 0.0 PARAMETER stop "<end_of_turn>" ``` However this produces outputs without tool calls
GiteaMirror added the questionmodel labels 2026-04-22 18:40:32 -05:00
GiteaMirror commented 2026-04-22 18:40:53 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 20, 2026):

ollama has a built-in renderer/parser for functiongemma, a template is not required.

FROM  functiongemma-finetune

RENDERER functiongemma
PARSER functiongemma

PARAMETER temperature 0.0
PARAMETER stop "<end_of_turn>"
<!-- gh-comment-id:3775446403 --> @rick-github commented on GitHub (Jan 20, 2026): ollama has a built-in renderer/parser for functiongemma, a template is not required. ```dockerfile FROM functiongemma-finetune RENDERER functiongemma PARSER functiongemma PARAMETER temperature 0.0 PARAMETER stop "<end_of_turn>" ```
GiteaMirror commented 2026-04-22 18:40:54 -05:00
Author
Owner
Copy Link

@derrick56007 commented on GitHub (Jan 21, 2026):

Thanks for that info, will give it a try!

<!-- gh-comment-id:3780864926 --> @derrick56007 commented on GitHub (Jan 21, 2026): Thanks for that info, will give it a try!
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 model 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#34802
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?