github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #5909] " Error: json: cannot unmarshal array into Go struct field Params.eos_token_id of type int " while importing llama 3.1 8B safetensor model from huggingface #3688

New Issue
Closed
opened 2026-04-12 14:30:14 -05:00 by GiteaMirror · 12 comments
GiteaMirror commented 2026-04-12 14:30:14 -05:00
Owner
Copy Link

Originally created by @sadegh-pz on GitHub (Jul 24, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5909

Originally assigned to: @joshyan1 on GitHub.

What is the issue?

What is the problem?

I was importing the llama 3.1 8B model from huggingface (meta-llama/Meta-Llama-3.1-8B-Instruct) using ollama create -f Modelfile but i got this error:
Error: json: cannot unmarshal array into Go struct field Params.eos_token_id of type int
I found the shallow cause of the error. and i fixed it naively. but i wanted to know why this error happned and what is the better solution.

What I get from the error?

the Go struct gets int for eos_token_id.
But the config.json file and generation_config.json file contain "eos_token_id": [128001, 128008, 128009].
So they face type mismatch and we get the error.

What is my naive solution?

I changed the eos_token_id value to "eos_token_id": 128001 without a concrete reason why i chose 128001 instead of the others. and it solved the problem and a model is created.

Why this error happened? What is the correct solution?

I don't know. I'd be appreciated to know.

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.2.8

Originally created by @sadegh-pz on GitHub (Jul 24, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5909 Originally assigned to: @joshyan1 on GitHub. ### What is the issue? ## What is the problem? I was importing the llama 3.1 8B model from huggingface (`meta-llama/Meta-Llama-3.1-8B-Instruct`) using `ollama create -f Modelfile` but i got this error: `Error: json: cannot unmarshal array into Go struct field Params.eos_token_id of type int` I found the shallow cause of the error. and i fixed it naively. but i wanted to know why this error happned and what is the better solution. ## What I get from the error? the `Go struct` gets `int` for `eos_token_id`. But the `config.json` file and `generation_config.json` file contain `"eos_token_id": [128001, 128008, 128009]`. So they face type mismatch and we get the error. ## What is my naive solution? I changed the `eos_token_id` value to `"eos_token_id": 128001` without a concrete reason why i chose `128001` instead of the others. and it solved the problem and a model is created. ## Why this error happened? What is the correct solution? I don't know. I'd be appreciated to know. ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.2.8
GiteaMirror added the bug label 2026-04-12 14:30:14 -05:00
GiteaMirror commented 2026-04-12 14:30:15 -05:00
Author
Owner
Copy Link

@ygivenx commented on GitHub (Jul 24, 2024):

I see this error in the docker logs

error from daemon in stream: Error grabbing logs: invalid character 'l' after object key:value pair

<!-- gh-comment-id:2248632974 --> @ygivenx commented on GitHub (Jul 24, 2024): I see this error in the docker logs `error from daemon in stream: Error grabbing logs: invalid character 'l' after object key:value pair`
GiteaMirror commented 2026-04-12 14:30:16 -05:00
Author
Owner
Copy Link

@mingmars commented on GitHub (Jul 25, 2024):

ollama create llama3.1 -f lama31.txt
transferring model data
unpacking model metadata
Error: json: cannot unmarshal array into Go struct field Params.eos_token_id of type int

配置文件是：

FROM D:\LLaMA-Factory\Meta-Llama-3.1-8B-Instruct
TEMPLATE "{{ if .System }}<|start_header_id|>system<|end_header_id|>

{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>

{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>

{{ .Response }}<|eot_id|>"
PARAMETER stop <|start_header_id|>
PARAMETER stop <|end_header_id|>
PARAMETER stop <|eot_id|>
PARAMETER num_keep 24

<!-- gh-comment-id:2250209684 --> @mingmars commented on GitHub (Jul 25, 2024): ollama create llama3.1 -f lama31.txt transferring model data unpacking model metadata Error: json: cannot unmarshal array into Go struct field Params.eos_token_id of type int 配置文件是： FROM D:\LLaMA-Factory\Meta-Llama-3.1-8B-Instruct TEMPLATE "{{ if .System }}<|start_header_id|>system<|end_header_id|> {{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|> {{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|> {{ .Response }}<|eot_id|>" PARAMETER stop <|start_header_id|> PARAMETER stop <|end_header_id|> PARAMETER stop <|eot_id|> PARAMETER num_keep 24
GiteaMirror commented 2026-04-12 14:30:16 -05:00
Author
Owner
Copy Link

@dgallardox commented on GitHub (Jul 27, 2024):

This is not totally relevant but I was getting the exact same error in Postman when making a POST request to /api/generate. It turns out it was caused by quotes in the stream value. E.g.,:

{
"model":"llama3",
"prompt": "once upon a time",
"stream": "false" // this causes a problem
}

VS.

{
"model":"llama3",
"prompt": "once upon a time",
"stream": false // this works
}

You possibly have a similar syntax error somewhere?
<!-- gh-comment-id:2253701389 --> @dgallardox commented on GitHub (Jul 27, 2024): This is not totally relevant but I was getting the exact same error in Postman when making a POST request to `/api/generate`. It turns out it was caused by quotes in the stream value. E.g.,: ``` { "model":"llama3", "prompt": "once upon a time", "stream": "false" // this causes a problem } VS. { "model":"llama3", "prompt": "once upon a time", "stream": false // this works } You possibly have a similar syntax error somewhere?
GiteaMirror commented 2026-04-12 14:30:16 -05:00
Author
Owner
Copy Link

@Dudu0831 commented on GitHub (Jul 29, 2024):

I also encountered the same problem, upgrading to 3.0 has not been resolved

<!-- gh-comment-id:2255306767 --> @Dudu0831 commented on GitHub (Jul 29, 2024): I also encountered the same problem, upgrading to 3.0 has not been resolved
GiteaMirror commented 2026-04-12 14:30:16 -05:00
Author
Owner
Copy Link

@202030481266 commented on GitHub (Jul 29, 2024):

same problem

<!-- gh-comment-id:2256035627 --> @202030481266 commented on GitHub (Jul 29, 2024): same problem
GiteaMirror commented 2026-04-12 14:30:16 -05:00
Author
Owner
Copy Link

@hgaong commented on GitHub (Aug 1, 2024):

same problem

<!-- gh-comment-id:2262224466 --> @hgaong commented on GitHub (Aug 1, 2024): same problem
GiteaMirror commented 2026-04-12 14:30:17 -05:00
Author
Owner
Copy Link

@hgaong commented on GitHub (Aug 1, 2024):

I changed the eos_token_id value to "eos_token_id": 128001 without a concrete reason why i chose 128001 instead of the others. and it solved the problem and a model is created.

<!-- gh-comment-id:2262225217 --> @hgaong commented on GitHub (Aug 1, 2024): I changed the eos_token_id value to "eos_token_id": 128001 without a concrete reason why i chose 128001 instead of the others. and it solved the problem and a model is created.
GiteaMirror commented 2026-04-12 14:30:17 -05:00
Author
Owner
Copy Link

@nebula2020 commented on GitHub (Aug 2, 2024):

Same issue I have faced. Switching to gguf deployment now.

<!-- gh-comment-id:2264876781 --> @nebula2020 commented on GitHub (Aug 2, 2024): Same issue I have faced. Switching to gguf deployment now.
GiteaMirror commented 2026-04-12 14:30:17 -05:00
Author
Owner
Copy Link

@sergei202 commented on GitHub (Aug 15, 2024):

I was tearing my hair out with this issue and finally got it working:

What fixed it for me is changing the eos_token_id in config.json to 128001, as @hgaong had suggested.
That got the model created, but then it wouldn't stop correctly and ended up output trash.

I added to add this to the Modelfile before calling ollama create and that fixed it!

PARAMETER stop <|start_header_id|>
PARAMETER stop <|end_header_id|>
PARAMETER stop <|eot_id|>
TEMPLATE [removed for readability]
<!-- gh-comment-id:2292152741 --> @sergei202 commented on GitHub (Aug 15, 2024): I was tearing my hair out with this issue and finally got it working: What fixed it for me is changing the `eos_token_id` in `config.json` to `128001`, as @hgaong had suggested. That got the model created, but then it wouldn't stop correctly and ended up output trash. I added to add this to the `Modelfile` before calling `ollama create` and that fixed it! ``` PARAMETER stop <|start_header_id|> PARAMETER stop <|end_header_id|> PARAMETER stop <|eot_id|> TEMPLATE [removed for readability] ```
GiteaMirror commented 2026-04-12 14:30:18 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Sep 2, 2024):

I'm going to go ahead and close this. You shouldn't need to specify the TEMPLATE as it should get autodetected.

<!-- gh-comment-id:2323560733 --> @pdevine commented on GitHub (Sep 2, 2024): I'm going to go ahead and close this. You shouldn't need to specify the `TEMPLATE` as it should get autodetected.
GiteaMirror commented 2026-04-12 14:30:18 -05:00
Author
Owner
Copy Link

@nofreewill42 commented on GitHub (Nov 12, 2024):

I had the error:

(.venv) nofreewill@nofreewill:~/FineTuneLLM$ ollama create mymodel -f Modelfile 
transferring model data 100% 
converting model 
Error: json: cannot unmarshal array into Go struct field .model.merges of type string

Modelfile content with the error:

FROM ./finetuned_llama_32_3b_gguf
PARAMETER stop [INST]
PARAMETER stop [/INST]
TEMPLATE "[INST] {{ .Prompt }} [/INST]"

Fixed model file:

FROM ./finetuned_llama_32_3b_gguf/unsloth.Q4_K_M.gguf
PARAMETER stop [INST]
PARAMETER stop [/INST]
TEMPLATE "[INST] {{ .Prompt }} [/INST]"

Fix highlighted:
FROM ./finetuned_llama_32_3b_gguf/unsloth.Q4_K_M.gguf

<!-- gh-comment-id:2470241545 --> @nofreewill42 commented on GitHub (Nov 12, 2024): I had the error: ``` (.venv) nofreewill@nofreewill:~/FineTuneLLM$ ollama create mymodel -f Modelfile transferring model data 100% converting model Error: json: cannot unmarshal array into Go struct field .model.merges of type string ``` Modelfile content with the error: ``` FROM ./finetuned_llama_32_3b_gguf PARAMETER stop [INST] PARAMETER stop [/INST] TEMPLATE "[INST] {{ .Prompt }} [/INST]" ``` Fixed model file: ``` FROM ./finetuned_llama_32_3b_gguf/unsloth.Q4_K_M.gguf PARAMETER stop [INST] PARAMETER stop [/INST] TEMPLATE "[INST] {{ .Prompt }} [/INST]" ``` Fix highlighted: FROM ./finetuned_llama_32_3b_gguf/**unsloth.Q4_K_M.gguf**
GiteaMirror commented 2026-04-12 14:30:19 -05:00
Author
Owner
Copy Link

@rtrad89 commented on GitHub (Nov 21, 2024):

I was having a similar error when using "context": 1 in the request, so I replaced it with "context": [1] and the error was fixed.

<!-- gh-comment-id:2491030820 --> @rtrad89 commented on GitHub (Nov 21, 2024): I was having a similar error when using `"context": 1` in the request, so I replaced it with `"context": [1]` and the error was fixed.
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#3688
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?