github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #1533] Ollama logs/serve should show fully templatized body that is sent to the LLM #835

New Issue
Closed
opened 2026-04-12 10:30:08 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 10:30:08 -05:00
Owner
Copy Link

Originally created by @scpedicini on GitHub (Dec 14, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1533

Can we get ollama serve to show the fully populated "templatized" input in the logs? (input + modelfile template). It's difficult to determine if the template syntax language is working as expected when building custom models. Currently, using the /api/chat endpoint only logs out that the endpoint was hit.

Logs:

ollama-gpu  | [GIN] 2023/12/14 - 23:05:37 | 200 |  2.626169805s |      172.18.0.1 | POST     "/api/chat"

This makes it difficult to analyze how things work.

For example, here is the official ollama Modelfile for mistral:7b-instruct:

# Modelfile generated by "ollama show"
# To build a new Modelfile based on this one, replace the FROM line with:
# FROM mistral:7b-instruct-q5_K_M

FROM /root/.ollama/models/blobs/sha256:dfe9a77f22f5a72d25dcf6a55b7401c41ef92e28f3a7ad7c6f6f6b651023df55
TEMPLATE """[INST] {{ .System }} {{ .Prompt }} [/INST]
"""
PARAMETER stop "[INST]"
PARAMETER stop "[/INST]"
PARAMETER stop "<<SYS>>"
PARAMETER stop "<</SYS>>"

It is difficult to grok how such a simple prompt handles a complex request with an array of messages[] coming from the chat API endpoint /api/chat

"messages": [
    {
        "role": "system",
        "content": "You are a helpful assistant. Always follow instructions."
    },        {
        "role": "assistant",
        "content": "How can I help you?"
    },
    {
        "role": "user",
        "content": "How can I fold myself in half so that I can qualify as a carry-on with American Airlines?"
    }
]
Originally created by @scpedicini on GitHub (Dec 14, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1533 Can we get `ollama serve` to show the fully populated "templatized" input in the logs? (input + modelfile template). It's difficult to determine if the template syntax language is working as expected when building custom models. Currently, using the `/api/chat` endpoint only logs out that the endpoint was hit. Logs: ```bash ollama-gpu | [GIN] 2023/12/14 - 23:05:37 | 200 | 2.626169805s | 172.18.0.1 | POST "/api/chat" ``` This makes it difficult to analyze how things work. For example, here is the official ollama Modelfile for `mistral:7b-instruct`: ``` # Modelfile generated by "ollama show" # To build a new Modelfile based on this one, replace the FROM line with: # FROM mistral:7b-instruct-q5_K_M FROM /root/.ollama/models/blobs/sha256:dfe9a77f22f5a72d25dcf6a55b7401c41ef92e28f3a7ad7c6f6f6b651023df55 TEMPLATE """[INST] {{ .System }} {{ .Prompt }} [/INST] """ PARAMETER stop "[INST]" PARAMETER stop "[/INST]" PARAMETER stop "<<SYS>>" PARAMETER stop "<</SYS>>" ``` It is difficult to grok how such a simple prompt handles a complex request with an array of `messages[]` coming from the chat API endpoint `/api/chat` ```json "messages": [ { "role": "system", "content": "You are a helpful assistant. Always follow instructions." }, { "role": "assistant", "content": "How can I help you?" }, { "role": "user", "content": "How can I fold myself in half so that I can qualify as a carry-on with American Airlines?" } ] ```
GiteaMirror added the feature request label 2026-04-12 10:30:08 -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 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#835
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?