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 #3974] error loading model architecture: unknown model architecture: 'phi3' #28217

New Issue
Closed
opened 2026-04-22 06:07:41 -05:00 by GiteaMirror · 12 comments
GiteaMirror commented 2026-04-22 06:07:41 -05:00
Owner
Copy Link

Originally created by @sanyuan0704 on GitHub (Apr 27, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3974

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

When i fine-tuned the phi3 and quantize it with latest llama.cpp, i found ollama cannot load the model:

image

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.1.32

Originally created by @sanyuan0704 on GitHub (Apr 27, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3974 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? When i fine-tuned the phi3 and quantize it with latest llama.cpp, i found ollama cannot load the model: ![image](https://github.com/ollama/ollama/assets/39261479/6c1cf96f-9bd7-4436-9b92-9d76537fde58) ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.1.32
GiteaMirror added the bug label 2026-04-22 06:07:41 -05:00
GiteaMirror commented 2026-04-22 06:07:43 -05:00
Author
Owner
Copy Link

@dotnetautor commented on GitHub (Apr 27, 2024):

+1

<!-- gh-comment-id:2080398772 --> @dotnetautor commented on GitHub (Apr 27, 2024): +1
GiteaMirror commented 2026-04-22 06:07:44 -05:00
Author
Owner
Copy Link

@johanteekens commented on GitHub (Apr 28, 2024):

+1

<!-- gh-comment-id:2081488512 --> @johanteekens commented on GitHub (Apr 28, 2024): +1
GiteaMirror commented 2026-04-22 06:07:45 -05:00
Author
Owner
Copy Link

@mildwood commented on GitHub (Apr 30, 2024):

same for me :(

<!-- gh-comment-id:2085248536 --> @mildwood commented on GitHub (Apr 30, 2024): same for me :(
GiteaMirror commented 2026-04-22 06:07:45 -05:00
Author
Owner
Copy Link

@fran-pastor commented on GitHub (Apr 30, 2024):

+1

<!-- gh-comment-id:2085785240 --> @fran-pastor commented on GitHub (Apr 30, 2024): +1
GiteaMirror commented 2026-04-22 06:07:46 -05:00
Author
Owner
Copy Link

@Phil-T1 commented on GitHub (May 1, 2024):

Me too.

<!-- gh-comment-id:2088495682 --> @Phil-T1 commented on GitHub (May 1, 2024): Me too.
GiteaMirror commented 2026-04-22 06:07:47 -05:00
Author
Owner
Copy Link

@kroni66 commented on GitHub (May 1, 2024):

Me too

<!-- gh-comment-id:2088705044 --> @kroni66 commented on GitHub (May 1, 2024): Me too
GiteaMirror commented 2026-04-22 06:07:48 -05:00
Author
Owner
Copy Link

@Phil-T1 commented on GitHub (May 1, 2024):

This is due to a recent change to the microsoft/Phi-3-mini-4k-instruct-gguf model apparently.

Apologies for not being specific as I'm on the move, but I replaced the Phi3 download I did today which gave the unknown Phi3 model architecture error, with a download from a couple of weeks ago, and the error went away.

I think Microsoft just changed the label from lama3 to phi3 in the latest version.

<!-- gh-comment-id:2088985676 --> @Phil-T1 commented on GitHub (May 1, 2024): This is due to a recent change to the microsoft/Phi-3-mini-4k-instruct-gguf model apparently. Apologies for not being specific as I'm on the move, but I replaced the Phi3 download I did today which gave the unknown Phi3 model architecture error, with a download from a couple of weeks ago, and the error went away. I think Microsoft just changed the label from lama3 to phi3 in the latest version.
GiteaMirror commented 2026-04-22 06:07:48 -05:00
Author
Owner
Copy Link

@robbiemu commented on GitHub (May 1, 2024):

this discussion may be tracking the resolution: https://github.com/ggerganov/llama.cpp/issues/6849

<!-- gh-comment-id:2088990826 --> @robbiemu commented on GitHub (May 1, 2024): this discussion may be tracking the resolution: https://github.com/ggerganov/llama.cpp/issues/6849
GiteaMirror commented 2026-04-22 06:07:49 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (May 1, 2024):

Initial support for phi3 was added in 0.1.32, and conversion should be working in 0.1.33. Please give the latest RC a try and let us know if you're still having problems.

https://github.com/ollama/ollama/releases

<!-- gh-comment-id:2089207020 --> @dhiltgen commented on GitHub (May 1, 2024): Initial support for phi3 was added in 0.1.32, and conversion should be working in 0.1.33. Please give the latest RC a try and let us know if you're still having problems. https://github.com/ollama/ollama/releases
GiteaMirror commented 2026-04-22 06:07:51 -05:00
Author
Owner
Copy Link

@MuazmaZ commented on GitHub (May 14, 2024):

+1

<!-- gh-comment-id:2109099432 --> @MuazmaZ commented on GitHub (May 14, 2024): +1
GiteaMirror commented 2026-04-22 06:07:52 -05:00
Author
Owner
Copy Link

@yorek commented on GitHub (May 14, 2024):

Hi, I've just tried latest version 0.1.37 and now I get the following error:

lama_model_load: error loading model: tensor 'blk.26.ffn_up.weight' data is not within the file bounds, model is corrupted or incomplete

here's the Modelfile I'm using:

FROM ./Phi-3-mini-4k-instruct-fp16.gguf
TEMPLATE """<s>{{ if .Prompt }}<|user|>
{{ .Prompt }}<|end|>
{{ end }}<|assistant|>
{{ .Response }}<|end|>"""
PARAMETER stop <|endoftext|>
PARAMETER stop <|assistant|>
PARAMETER stop <|end|>
PARAMETER num_ctx 4096

the GGUF file is coming from the Hugging Face MS repo:

https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-gguf/tree/main

Thanks

<!-- gh-comment-id:2109110531 --> @yorek commented on GitHub (May 14, 2024): Hi, I've just tried latest version 0.1.37 and now I get the following error: ``` lama_model_load: error loading model: tensor 'blk.26.ffn_up.weight' data is not within the file bounds, model is corrupted or incomplete ``` here's the Modelfile I'm using: ``` FROM ./Phi-3-mini-4k-instruct-fp16.gguf TEMPLATE """<s>{{ if .Prompt }}<|user|> {{ .Prompt }}<|end|> {{ end }}<|assistant|> {{ .Response }}<|end|>""" PARAMETER stop <|endoftext|> PARAMETER stop <|assistant|> PARAMETER stop <|end|> PARAMETER num_ctx 4096 ``` the GGUF file is coming from the Hugging Face MS repo: https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-gguf/tree/main Thanks
GiteaMirror commented 2026-04-22 06:07:52 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Jun 4, 2024):

Hi all this should be fixed now. Sorry about the errors - let me know if you're still seeing it on the latest version of Ollama

<!-- gh-comment-id:2146736176 --> @jmorganca commented on GitHub (Jun 4, 2024): Hi all this should be fixed now. Sorry about the errors - let me know if you're still seeing it on the latest version of Ollama
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#28217
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?