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 #9580] Getting gathering model components Error: invalid model name when trying to create model from local GGUF #6249

New Issue
Closed
opened 2026-04-12 17:40:59 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 17:40:59 -05:00
Owner
Copy Link

Originally created by @phiwi on GitHub (Mar 7, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9580

What is the issue?

Ollama version: 0.5.13

I have a Modelfile located at ~/model_cards/Modelfile that points to a finetuned model with huggingface & unsloth. Inside the Modelfile:

FROM /xxx/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/unsloth.Q8_0.gguf

The model directory /prj/LINDA_LLM/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/ looks the following (there is also a Modelfile, but it doesn't matter if I use this or the above self-created one):

├── config.json
├── generation_config.json
├── model-00001-of-00004.safetensors
├── model-00002-of-00004.safetensors
├── model-00003-of-00004.safetensors
├── model-00004-of-00004.safetensors
├── Modelfile
├── model.safetensors.index.json
├── special_tokens_map.json
├── tokenizer_config.json
├── tokenizer.json
└── unsloth.Q8_0.gguf

Now, when running

ollama create -f ./Modelfile  llama3.1-128k-regu:8b

or

ollama create -f Modelfile  llama3.1-128k-regu:8b

inside ~/model_cards/

I get

gathering model components 
Error: invalid model name

I also tried with the pre-made Modelfile from unsloth - same result.

I changed the command FROM ... to

  • /xxx/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/ (just the folder)
  • .../outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/ (relative pathing)
  • /xxx/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B_merged_16_bit/ (folder with just safetensors)

and same result.

Relevant log output

gathering model components 
Error: invalid model name

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @phiwi on GitHub (Mar 7, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9580 ### What is the issue? Ollama version: `0.5.13` I have a `Modelfile` located at `~/model_cards/Modelfile` that points to a finetuned model with huggingface & unsloth. Inside the Modelfile: ``` FROM /xxx/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/unsloth.Q8_0.gguf ``` The model directory `/prj/LINDA_LLM/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/` looks the following (there is also a Modelfile, but it doesn't matter if I use this or the above self-created one): ``` ├── config.json ├── generation_config.json ├── model-00001-of-00004.safetensors ├── model-00002-of-00004.safetensors ├── model-00003-of-00004.safetensors ├── model-00004-of-00004.safetensors ├── Modelfile ├── model.safetensors.index.json ├── special_tokens_map.json ├── tokenizer_config.json ├── tokenizer.json └── unsloth.Q8_0.gguf ``` Now, when running ``` ollama create -f ./Modelfile llama3.1-128k-regu:8b ``` or ``` ollama create -f Modelfile llama3.1-128k-regu:8b ``` inside `~/model_cards/` I get ``` gathering model components Error: invalid model name ``` I also tried with the pre-made Modelfile from unsloth - same result. I changed the command `FROM ...` to - `/xxx/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/` (just the folder) - `.../outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B.GGUF/` (relative pathing) - `/xxx/outputs/finetunedmodels/unsloth/Meta-Llama-3.1-8B_merged_16_bit/` (folder with just safetensors) and same result. ### Relevant log output ```shell gathering model components Error: invalid model name ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-12 17:40:59 -05:00
GiteaMirror commented 2026-04-12 17:41:00 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 7, 2025):

You can try tracing the program execution to see if it's actually trying to access the files:

strace  -f --trace=newfstatat --signal='!SIGURG' ollama create -f ~/model_cards/Modelfile  llama3.1-128k-regu:8b

The other common reason for the invalid model name error is that the name of the new model is failing a check, eg too many ':'s (llama3.1-128k:regu:8b). But the model name you've chosen looks fine. Unless maybe there's some embedded UTF-8 characters that are not being rendered in the post.

<!-- gh-comment-id:2706776946 --> @rick-github commented on GitHub (Mar 7, 2025): You can try tracing the program execution to see if it's actually trying to access the files: ``` strace -f --trace=newfstatat --signal='!SIGURG' ollama create -f ~/model_cards/Modelfile llama3.1-128k-regu:8b ``` The other common reason for the `invalid model name` error is that the name of the new model is failing a check, eg too many ':'s (llama3.1-128k:regu:8b). But the model name you've chosen looks fine. Unless maybe there's some embedded UTF-8 characters that are not being rendered in the post.
GiteaMirror commented 2026-04-12 17:41:00 -05:00
Author
Owner
Copy Link

@phiwi commented on GitHub (Mar 10, 2025):

I couls solve it, although the error message was pretty misleading. In my case, as I was using Ollama via singularity, I missed to bind the paths both of the model file and the actual model.

<!-- gh-comment-id:2710976418 --> @phiwi commented on GitHub (Mar 10, 2025): I couls solve it, although the error message was pretty misleading. In my case, as I was using Ollama via singularity, I missed to bind the paths both of the model file and the actual model.
GiteaMirror commented 2026-04-12 17:41:01 -05:00
Author
Owner
Copy Link

@divinehui commented on GitHub (Mar 27, 2025):

I also meet this problem:
root@49fb19dc80a4:~/.ollama/models/qwq32b# ollama create -f /root/.ollama/models/qwq32b/Modelfile qwq:32b
gathering model components
copying file sha256:e242184c48437802c44fee7defc33dc26f5d121c72698bce2baf4c706d791c47 100%
Error: path or modelfile are required

The Modelfile as blow:
FROM /root/.ollama/models/qwq32b/qwq-32b-q5_0.gguf

TEMPLATE """{{- if .System }}
<|im_start|>system {{ .System }}<|im_end|>
{{- end }}
<|im_start|>user
{{ .Prompt }}<|im_end|>
<|im_start|>assistant
"""

SYSTEM """"""

PARAMETER stop <|im_start|>
PARAMETER stop <|im_end|>

How to solve this problem? anyone can help me?

<!-- gh-comment-id:2756674150 --> @divinehui commented on GitHub (Mar 27, 2025): I also meet this problem: root@49fb19dc80a4:~/.ollama/models/qwq32b# ollama create -f /root/.ollama/models/qwq32b/Modelfile qwq:32b gathering model components copying file sha256:e242184c48437802c44fee7defc33dc26f5d121c72698bce2baf4c706d791c47 100% Error: path or modelfile are required The Modelfile as blow: FROM /root/.ollama/models/qwq32b/qwq-32b-q5_0.gguf TEMPLATE """{{- if .System }} <|im_start|>system {{ .System }}<|im_end|> {{- end }} <|im_start|>user {{ .Prompt }}<|im_end|> <|im_start|>assistant """ SYSTEM """""" PARAMETER stop <|im_start|> PARAMETER stop <|im_end|> How to solve this problem? anyone can help me?
GiteaMirror commented 2026-04-12 17:41:01 -05:00
Author
Owner
Copy Link

@SamuelLarkin commented on GitHub (May 8, 2025):

I faced the same issue.
It looks like ollama create -f wants a file that is in the current directory.

ollama create --file Unbabel/TowerInstruct-7B-v0.1/sft/lora/en2fr/123581/merged/ollama.Modelfile Unbabel--TowerInstruct-7B-v0.1-f16.ollama

gathering model components
Error: invalid model name

Moved the Modelfile in the current directory.

cp Unbabel/TowerInstruct-7B-v0.1/sft/lora/en2fr/123581/merged/ollama.Modelfile .

Now it works

ollama create --file ollama.Modelfile Unbabel--TowerInstruct-7B-v0.1-f16

gathering model components
copying file sha256:0d8a2dbde1f27637b336424090db1ac5854233d63888fe46e3ebe5113192e26b 100%
parsing GGUF
using existing layer sha256:0d8a2dbde1f27637b336424090db1ac5854233d63888fe46e3ebe5113192e26b
using existing layer sha256:a602670d1b49da353a389dffff4b771933a46be7f2162267404808095b9fb9a1
writing manifest
success

ollama --version
ollama version is 0.5.7

<!-- gh-comment-id:2863408598 --> @SamuelLarkin commented on GitHub (May 8, 2025): I faced the same issue. It looks like `ollama create -f` wants a file that is in the current directory. ```sh ollama create --file Unbabel/TowerInstruct-7B-v0.1/sft/lora/en2fr/123581/merged/ollama.Modelfile Unbabel--TowerInstruct-7B-v0.1-f16.ollama ``` ``` gathering model components Error: invalid model name ``` Moved the `Modelfile` in the current directory. ```sh cp Unbabel/TowerInstruct-7B-v0.1/sft/lora/en2fr/123581/merged/ollama.Modelfile . ``` Now it works ```sh ollama create --file ollama.Modelfile Unbabel--TowerInstruct-7B-v0.1-f16 ``` ``` gathering model components copying file sha256:0d8a2dbde1f27637b336424090db1ac5854233d63888fe46e3ebe5113192e26b 100% parsing GGUF using existing layer sha256:0d8a2dbde1f27637b336424090db1ac5854233d63888fe46e3ebe5113192e26b using existing layer sha256:a602670d1b49da353a389dffff4b771933a46be7f2162267404808095b9fb9a1 writing manifest success ``` `ollama --version` ollama version is 0.5.7
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#6249
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?