github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-09 01:59:50 -05:00
Code Issues 14.6k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #6858] Unable to load adapter_model.safetensors for Phi3-Medium-128k #30088

New Issue
Closed
opened 2026-04-22 09:33:09 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 09:33:09 -05:00
Owner
Copy Link

Originally created by @AAndersn on GitHub (Sep 18, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6858

What is the issue?

Trying to load a safetensors adapter file for phi3-medium-128k using a .modelfile. I generated an adapter_config.json and adapter_model.safetensors files using lora training and copied them into the ollama docker container.

I have generated the modelfile with ollama show phi3:medium --modelfile > phi3_med_cim.modelfile and edited it to add

# Modelfile generated by "ollama show"
FROM phi3:medium
ADAPTER /home/medium_128k
TEMPLATE "{{ if .System }}<|system|>
...

When trying to create the model, I am calling ollama create phi3_med_cim --file /home/phi3_med_cim.modelfile, which throws the following error:

converting model 
Error: unsupported architecture

These steps work fine for phi3-mini-4k and llama3.1, but not for phi3-medium

OS

WSL2

GPU

Nvidia

CPU

Intel

Ollama version

0.3.10 (docker)

Originally created by @AAndersn on GitHub (Sep 18, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6858 ### What is the issue? Trying to load a safetensors adapter file for phi3-medium-128k using a .modelfile. I generated an adapter_config.json and adapter_model.safetensors files using lora training and copied them into the ollama docker container. I have generated the modelfile with `ollama show phi3:medium --modelfile > phi3_med_cim.modelfile` and edited it to add ``` # Modelfile generated by "ollama show" FROM phi3:medium ADAPTER /home/medium_128k TEMPLATE "{{ if .System }}<|system|> ... ``` When trying to create the model, I am calling `ollama create phi3_med_cim --file /home/phi3_med_cim.modelfile`, which throws the following error: ```transferring model data 100% converting model Error: unsupported architecture ``` These steps work fine for phi3-mini-4k and llama3.1, but not for phi3-medium ### OS WSL2 ### GPU Nvidia ### CPU Intel ### Ollama version 0.3.10 (docker)
GiteaMirror added the bug label 2026-04-22 09:33:09 -05:00
GiteaMirror commented 2026-04-22 09:33:14 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 19, 2024):

The model conversion in ollama supports a limted set of architectures. You may make more progress by using llama.cpp to convert the base model to GGUF and using that in your FROM line.

<!-- gh-comment-id:2359701121 --> @rick-github commented on GitHub (Sep 19, 2024): The model conversion in ollama supports a limted set of architectures. You may make more progress by using llama.cpp to [convert the base model](https://github.com/ollama/ollama/issues/6628#issuecomment-2329341148) to GGUF and using that in your `FROM` line.
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#30088
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?