github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 13:19:40 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #9783] Request for Modelfile for gemma-3-27b-it-Q4_K_M.gguf #6397

New Issue
Closed
opened 2026-04-12 17:55:19 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 17:55:19 -05:00
Owner
Copy Link

Originally created by @ZimaBlueee on GitHub (Mar 15, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9783

Hi, I'm new to Ollama and trying to use the gemma-3-27b-it-Q4_K_M.gguf model.

Could someone please provide a working Modelfile for this specific model? I’d like to ensure it’s configured correctly for optimal performance with Ollama.

Any additional tips for a beginner would also be greatly appreciated!

Thank you!

Originally created by @ZimaBlueee on GitHub (Mar 15, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9783 Hi, I'm new to Ollama and trying to use the gemma-3-27b-it-Q4_K_M.gguf model. Could someone please provide a working Modelfile for this specific model? I’d like to ensure it’s configured correctly for optimal performance with Ollama. Any additional tips for a beginner would also be greatly appreciated! Thank you!
GiteaMirror commented 2026-04-12 17:55:21 -05:00
Author
Owner
Copy Link

@svkoh commented on GitHub (Mar 15, 2025):

params and template are from https://ollama.com/library/gemma3

if you have less than 24G memory, reduce num_ctx for better performance.

FROM /path/to/gemma-3-27b-it-Q4_K_M.gguf

PARAMETER num_ctx 8192
PARAMETER temperature 0.1
PARAMETER stop "<end_of_turn>"

TEMPLATE """
{{- range $i, $_ := .Messages }}
{{- $last := eq (len (slice $.Messages $i)) 1 }}
{{- if or (eq .Role "user") (eq .Role "system") }}<start_of_turn>user
{{ .Content }}<end_of_turn>
{{ if $last }}<start_of_turn>model
{{ end }}
{{- else if eq .Role "assistant" }}<start_of_turn>model
{{ .Content }}{{ if not $last }}<end_of_turn>
{{ end }}
{{- end }}
{{- end }}
"""
<!-- gh-comment-id:2726499020 --> @svkoh commented on GitHub (Mar 15, 2025): params and template are from [https://ollama.com/library/gemma3](https://ollama.com/library/gemma3) if you have less than 24G memory, reduce num_ctx for better performance. ``` FROM /path/to/gemma-3-27b-it-Q4_K_M.gguf PARAMETER num_ctx 8192 PARAMETER temperature 0.1 PARAMETER stop "<end_of_turn>" TEMPLATE """ {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice $.Messages $i)) 1 }} {{- if or (eq .Role "user") (eq .Role "system") }}<start_of_turn>user {{ .Content }}<end_of_turn> {{ if $last }}<start_of_turn>model {{ end }} {{- else if eq .Role "assistant" }}<start_of_turn>model {{ .Content }}{{ if not $last }}<end_of_turn> {{ end }} {{- end }} {{- end }} """ ```
GiteaMirror commented 2026-04-12 17:55:22 -05:00
Author
Owner
Copy Link

@ZimaBlueee commented on GitHub (Mar 15, 2025):

Thank you!

<!-- gh-comment-id:2726551438 --> @ZimaBlueee commented on GitHub (Mar 15, 2025): Thank you!
GiteaMirror commented 2026-04-12 17:55:23 -05:00
Author
Owner
Copy Link

@wqerrewetw commented on GitHub (Mar 18, 2025):

template with tool support

{{- if .Messages }}
{{- if or .System .Tools }}
{{- if .System }}
{{ .System }}
{{- end }}
{{- if .Tools }}
The following tools are available when needed for specific tasks:
{{ .Tools }}

Only use tools when the task specifically requires their functionality.
For general questions or tasks that don't need external data, respond directly.

When using a tool, format as:
<tool_calls>
<tool_call>
{"name": "function_name", "parameters": {"param1": "value1"}}
</tool_call>
</tool_calls>
{{- end }}
{{- end }}
{{- range $i, $_ := .Messages }}
{{- $last := eq (len (slice $.Messages $i)) 1 }}
{{- if eq .Role "user" }}<start_of_turn>user
{{ .Content }}<end_of_turn>
{{- if $last }}<start_of_turn>model
{{ end }}
{{- else if eq .Role "system" }}<start_of_turn>user
{{ .Content }}<end_of_turn>
{{- if $last }}<start_of_turn>model
{{ end }}
{{- else if eq .Role "assistant" }}<start_of_turn>model
{{- if .ToolCalls }}
<tool_calls>
{{- range .ToolCalls }}
<tool_call>
{"name": "{{ .Function.Name }}", "parameters": {{ .Function.Arguments }}}
</tool_call>
{{- end }}
</tool_calls>
{{- else }}
{{ .Content }}
{{- end }}
{{- if not $last }}<end_of_turn>
{{ end }}
{{- else if eq .Role "tool" }}
<tool_outputs>
<tool_output>
{{ .Content }}
</tool_output>
</tool_outputs>
{{- if and $last (ne .Role "assistant") }}<start_of_turn>model
{{- end }}
{{- end }}
{{- end }}
{{- else }}
{{- if .System }}
{{ .System }}
{{- end }}
{{- if .Prompt }}
<start_of_turn>user
{{ .Prompt }}<end_of_turn>
<start_of_turn>model
{{- end }}
{{ .Response }}
{{- if .Response }}{{ end }}
{{- end }}
<!-- gh-comment-id:2731353986 --> @wqerrewetw commented on GitHub (Mar 18, 2025): template with tool support ``` {{- if .Messages }} {{- if or .System .Tools }} {{- if .System }} {{ .System }} {{- end }} {{- if .Tools }} The following tools are available when needed for specific tasks: {{ .Tools }} Only use tools when the task specifically requires their functionality. For general questions or tasks that don't need external data, respond directly. When using a tool, format as: <tool_calls> <tool_call> {"name": "function_name", "parameters": {"param1": "value1"}} </tool_call> </tool_calls> {{- end }} {{- end }} {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice $.Messages $i)) 1 }} {{- if eq .Role "user" }}<start_of_turn>user {{ .Content }}<end_of_turn> {{- if $last }}<start_of_turn>model {{ end }} {{- else if eq .Role "system" }}<start_of_turn>user {{ .Content }}<end_of_turn> {{- if $last }}<start_of_turn>model {{ end }} {{- else if eq .Role "assistant" }}<start_of_turn>model {{- if .ToolCalls }} <tool_calls> {{- range .ToolCalls }} <tool_call> {"name": "{{ .Function.Name }}", "parameters": {{ .Function.Arguments }}} </tool_call> {{- end }} </tool_calls> {{- else }} {{ .Content }} {{- end }} {{- if not $last }}<end_of_turn> {{ end }} {{- else if eq .Role "tool" }} <tool_outputs> <tool_output> {{ .Content }} </tool_output> </tool_outputs> {{- if and $last (ne .Role "assistant") }}<start_of_turn>model {{- end }} {{- end }} {{- end }} {{- else }} {{- if .System }} {{ .System }} {{- end }} {{- if .Prompt }} <start_of_turn>user {{ .Prompt }}<end_of_turn> <start_of_turn>model {{- end }} {{ .Response }} {{- if .Response }}{{ end }} {{- end }} ```
GiteaMirror commented 2026-04-12 17:55:24 -05:00
Author
Owner
Copy Link

@svkoh commented on GitHub (May 20, 2025):

This tool can convert huggingface chat template to the Go format used by Ollama: @huggingface/ollama-utils

Online version: https://huggingface.co/spaces/ngxson/debug_ollama_manifest

<!-- gh-comment-id:2893515750 --> @svkoh commented on GitHub (May 20, 2025): This tool can convert huggingface chat template to the Go format used by Ollama: [@huggingface/ollama-utils](https://github.com/huggingface/huggingface.js/blob/main/packages/ollama-utils/README.md) Online version: [https://huggingface.co/spaces/ngxson/debug_ollama_manifest](https://huggingface.co/spaces/ngxson/debug_ollama_manifest)
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#6397
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?