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 #10315] Local Model Not Working Properly and Ollama Keeps Asking to Pull #68830

New Issue
Closed
opened 2026-05-04 15:22:36 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-04 15:22:36 -05:00
Owner
Copy Link

Originally created by @speechless-z on GitHub (Apr 17, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10315

What is the issue?

❓ Problem: Local Model Not Working Properly and Ollama Keeps Asking to Pull
Hi! I'm trying to run deepseek-r1 on a computer that cannot connect to the internet, and I've encountered a few issues:

🧩 My Setup:
On an online computer:

I successfully downloaded Ollama and the deepseek-r1 model.

I copied the blobs and manifest files to the offline computer.

I also ran ollama show to generate a Modelfile.

On the offline computer:

I used the command:

bash
ollama create deepseek-r17b.gguf -f Modelfile

My Modelfile looks like this:

FROM F:\OllamaModels\blobs\sha256-96c415656d377afbff962f6cdb2394ab092ccbcbaab4b82525bc4ca800fe8a49
TEMPLATE """{{- if .System }}{{ .System }}{{ end }}
{{- range $i, $_ := .Messages }}
{{- last := eq (len (slice .Messages $i)) 1}}
{{- if eq .Role "user" }}<｜User｜>{{ .Content }}
{{- else if eq .Role "assistant" }}<｜Assistant｜>{{ .Content }}{{- if not $last }}<｜end▁of▁sentence｜>{{- end }}
{{- end }}
{{- if and $last (ne .Role "assistant") }}<｜Assistant｜>{{- end }}
{{- end }}"""
PARAMETER stop <｜begin▁of▁sentence｜>
PARAMETER stop <｜end▁of▁sentence｜>
PARAMETER stop <｜User｜>
PARAMETER stop <｜Assistant｜>
LICENSE """MIT License ... (DeepSeek license)
"""
😫 Issues I'm Facing:
The model outputs nonsense or just empty responses
Even though it technically runs, it often outputs nothing, question marks (?), or irrelevant gibberish.

Ollama keeps trying to pull the model
Every time I restart ollama and try to run the model again, it prompts me to pull it online — but the machine has no internet access. How can I make it recognize and use the local model permanently?

🧠 What I Need:
How can I correctly use a downloaded model (from blobs) without internet?

How do I make ollama recognize the model as already installed, so it doesn't attempt to pull it?

Do I need to adjust the FROM line in the Modelfile or move files to a specific folder?

Any suggestions on fixing the model hallucination/empty response issue?

Thanks in advance! 🙏

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @speechless-z on GitHub (Apr 17, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10315 ### What is the issue? ❓ Problem: Local Model Not Working Properly and Ollama Keeps Asking to Pull Hi! I'm trying to run deepseek-r1 on a computer that cannot connect to the internet, and I've encountered a few issues: 🧩 My Setup: On an online computer: I successfully downloaded Ollama and the deepseek-r1 model. I copied the blobs and manifest files to the offline computer. I also ran ollama show to generate a Modelfile. On the offline computer: I used the command: bash ollama create deepseek-r17b.gguf -f Modelfile My Modelfile looks like this: FROM F:\OllamaModels\blobs\sha256-96c415656d377afbff962f6cdb2394ab092ccbcbaab4b82525bc4ca800fe8a49 TEMPLATE """{{- if .System }}{{ .System }}{{ end }} {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice $.Messages $i)) 1}} {{- if eq .Role "user" }}<｜User｜>{{ .Content }} {{- else if eq .Role "assistant" }}<｜Assistant｜>{{ .Content }}{{- if not $last }}<｜end▁of▁sentence｜>{{- end }} {{- end }} {{- if and $last (ne .Role "assistant") }}<｜Assistant｜>{{- end }} {{- end }}""" PARAMETER stop <｜begin▁of▁sentence｜> PARAMETER stop <｜end▁of▁sentence｜> PARAMETER stop <｜User｜> PARAMETER stop <｜Assistant｜> LICENSE """MIT License ... (DeepSeek license) """ 😫 Issues I'm Facing: The model outputs nonsense or just empty responses Even though it technically runs, it often outputs nothing, question marks (?), or irrelevant gibberish. Ollama keeps trying to pull the model Every time I restart ollama and try to run the model again, it prompts me to pull it online — but the machine has no internet access. How can I make it recognize and use the local model permanently? 🧠 What I Need: How can I correctly use a downloaded model (from blobs) without internet? How do I make ollama recognize the model as already installed, so it doesn't attempt to pull it? Do I need to adjust the FROM line in the Modelfile or move files to a specific folder? Any suggestions on fixing the model hallucination/empty response issue? Thanks in advance! 🙏 ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bugneeds more info labels 2026-05-04 15:22:37 -05:00
GiteaMirror commented 2026-05-04 15:22:41 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 17, 2025):

Every time I restart ollama and try to run the model again, it prompts me to pull it online

If ollama wants to pull the model when you run it, how are you getting any responses?

How can I correctly use a downloaded model (from blobs) without internet?

The procedure you've outlined should work: copy the model blob to the computer, create a Modelfile, run ollama create. However, you have called the model deepseek-r17b.gguf, rather than the usual deepseek-r1:7b. Perhaps that is the issue, the wrong name is being used. What's the output of ollama list?

How do I make ollama recognize the model as already installed, so it doesn't attempt to pull it?

Once you've done ollama create, the model is installed.

Do I need to adjust the FROM line in the Modelfile or move files to a specific folder?

If the ollama create didn't emit an error message, the Modelfile is fine.

Any suggestions on fixing the model hallucination/empty response issue?

The model should perform the same as elsewhere, so perhaps an environmental issue. Server logs may aid in debugging.

<!-- gh-comment-id:2812384079 --> @rick-github commented on GitHub (Apr 17, 2025): > Every time I restart ollama and try to run the model again, it prompts me to pull it online If ollama wants to pull the model when you run it, how are you getting any responses? > How can I correctly use a downloaded model (from blobs) without internet? The procedure you've outlined should work: copy the model blob to the computer, create a Modelfile, run `ollama create`. However, you have called the model `deepseek-r17b.gguf`, rather than the usual `deepseek-r1:7b`. Perhaps that is the issue, the wrong name is being used. What's the output of `ollama list`? > How do I make ollama recognize the model as already installed, so it doesn't attempt to pull it? Once you've done `ollama create`, the model is installed. > Do I need to adjust the FROM line in the Modelfile or move files to a specific folder? If the `ollama create` didn't emit an error message, the Modelfile is fine. > Any suggestions on fixing the model hallucination/empty response issue? The model should perform the same as elsewhere, so perhaps an environmental issue. [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may aid in debugging.
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 needs more info
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#68830
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?