github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #6968] Adjust templates for FIM models to acknowledge existence of suffix #66457

New Issue
Open
opened 2026-05-04 05:35:32 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-05-04 05:35:32 -05:00
Owner
Copy Link

Originally created by @sestinj on GitHub (Sep 25, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6968

What is the issue?

CodeGemma (for example, it's not the only one) supports both FIM and chat. Ollama uses the FIM template for codegemma:2b and the chat template for codegemma:7b. This feels like the right default decision, but in cases where a suffix is provided, it can be confidently assumed that a FIM format is desired.

The solution to this would be a prompt template that is wrapped in an "if suffix", something like this:

{{- if .Suffix }}<|fim_prefix|>{{ .Prompt }}<|fim_suffix|>{{ .Suffix }}<|fim_middle|>
{{- else }}<start_of_turn>user
{{ if .System }}{{ .System }} {{ end }}{{ .Prompt }}<end_of_turn>
<start_of_turn>model
{{ .Response }}<end_of_turn>
{{- end }}

One caveat I can see is it would make the prompt template slightly less readable for users browsing the Ollama template library.

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.3.11

Originally created by @sestinj on GitHub (Sep 25, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6968 ### What is the issue? CodeGemma (for example, it's not the only one) supports both FIM and chat. Ollama uses the FIM template for codegemma:2b and the chat template for codegemma:7b. This feels like the right default decision, but in cases where a suffix is provided, it can be confidently assumed that a FIM format is desired. The solution to this would be a prompt template that is wrapped in an "if suffix", something like this: ``` {{- if .Suffix }}<|fim_prefix|>{{ .Prompt }}<|fim_suffix|>{{ .Suffix }}<|fim_middle|> {{- else }}<start_of_turn>user {{ if .System }}{{ .System }} {{ end }}{{ .Prompt }}<end_of_turn> <start_of_turn>model {{ .Response }}<end_of_turn> {{- end }} ``` One caveat I can see is it would make the prompt template slightly less readable for users browsing the Ollama template library. ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.3.11
GiteaMirror added the bug label 2026-05-04 05:35:32 -05:00
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 bug
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#66457
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?