github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 08:59:13 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #13017] Bug: 500 Internal Server Error: unmarshal: invalid character 'I' with 'kimi-k2-thinking:cloud' model #70678

New Issue
Closed
opened 2026-05-04 22:31:21 -05:00 by GiteaMirror · 12 comments
GiteaMirror commented 2026-05-04 22:31:21 -05:00
Owner
Copy Link

Originally created by @TJbarros on GitHub (Nov 8, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13017

What is the issue?

Summary
This is a bug report for the kimi-k2-thinking:cloud model endpoint. The ollama.com server is consistently returning a 500 Internal Server Error.

This error response is not valid JSON, which causes the ollama client to fail with an unmarshal error.

🚶 Steps to Reproduce
Run the kimi-k2-thinking:cloud model:

Bash

ollama run kimi-k2-thinking:cloud
After connecting, send any prompt. (I used qual a sua capacidade de programar em python, seja cincero e objetivo ?, but any prompt triggers the error).

💥 Actual Behavior (The Error)
The client immediately prints the following error, indicating a non-JSON response was received from the server.

Connecting to 'kimi-k2-thinking' on 'ollama.com' ⚡

qual a sua capacidade de programar em python, seja cincero e objetivo ?
Error: 500 Internal Server Error: unmarshal: invalid character 'I' looking for beginning of value
🎯 Error Analysis
The key part of the error is: unmarshal: invalid character 'I' looking for beginning of value.

This confirms that the ollama client received a 500 Internal Server Error and the body of that error response was not JSON. The client expected a JSON object (starting with { or [) but instead received a response body starting with the character I.

This is almost certainly a plain-text or HTML error page from the server (e.g., "Internal Server Error" or "Invalid...").

This suggests the kimi-k2-thinking:cloud endpoint is failing and not correctly formatting its error response as a JSON object that the client can understand.

Environment
Ollama Version: ollama version is 0.12.10
OS: Windows / PowerShell

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @TJbarros on GitHub (Nov 8, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13017 ### What is the issue? Summary This is a bug report for the kimi-k2-thinking:cloud model endpoint. The ollama.com server is consistently returning a 500 Internal Server Error. This error response is not valid JSON, which causes the ollama client to fail with an unmarshal error. 🚶 Steps to Reproduce Run the kimi-k2-thinking:cloud model: Bash ollama run kimi-k2-thinking:cloud After connecting, send any prompt. (I used qual a sua capacidade de programar em python, seja cincero e objetivo ?, but any prompt triggers the error). 💥 Actual Behavior (The Error) The client immediately prints the following error, indicating a non-JSON response was received from the server. Connecting to 'kimi-k2-thinking' on 'ollama.com' ⚡ >>> qual a sua capacidade de programar em python, seja cincero e objetivo ? Error: 500 Internal Server Error: unmarshal: invalid character 'I' looking for beginning of value 🎯 Error Analysis The key part of the error is: unmarshal: invalid character 'I' looking for beginning of value. This confirms that the ollama client received a 500 Internal Server Error and the body of that error response was not JSON. The client expected a JSON object (starting with { or [) but instead received a response body starting with the character I. This is almost certainly a plain-text or HTML error page from the server (e.g., "Internal Server Error" or "Invalid..."). This suggests the kimi-k2-thinking:cloud endpoint is failing and not correctly formatting its error response as a JSON object that the client can understand. Environment Ollama Version: ollama version is 0.12.10 OS: Windows / PowerShell ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-05-04 22:31:21 -05:00
GiteaMirror commented 2026-05-04 22:31:25 -05:00
Author
Owner
Copy Link

@benhot commented on GitHub (Nov 8, 2025):

same issues

Image
<!-- gh-comment-id:3506556567 --> @benhot commented on GitHub (Nov 8, 2025): same issues <img width="547" height="505" alt="Image" src="https://github.com/user-attachments/assets/1ed2ea6d-30eb-4308-9a22-75e854c764f7" />
GiteaMirror commented 2026-05-04 22:31:26 -05:00
Author
Owner
Copy Link

@josephlugo commented on GitHub (Nov 8, 2025):

Yep

<!-- gh-comment-id:3506562129 --> @josephlugo commented on GitHub (Nov 8, 2025): <!-- Failed to upload "IMG_1137.jpeg" --> Yep
GiteaMirror commented 2026-05-04 22:31:27 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 8, 2025):

This should be fixed now.

$ ollama run kimi-k2-thinking:cloud hello
...
Hello! How can I help you today?
<!-- gh-comment-id:3506790483 --> @rick-github commented on GitHub (Nov 8, 2025): This should be fixed now. ```console $ ollama run kimi-k2-thinking:cloud hello ... Hello! How can I help you today? ```
GiteaMirror commented 2026-05-04 22:31:28 -05:00
Author
Owner
Copy Link

@benhot commented on GitHub (Nov 9, 2025):

fixed

<!-- gh-comment-id:3507535524 --> @benhot commented on GitHub (Nov 9, 2025): fixed
GiteaMirror commented 2026-05-04 22:31:29 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 9, 2025):

It appears this is an intermittent ongoing issue. The invalid character 'I' is a backend error that's causing Internal Server Error to be returned, and because it's not properly formed JSON, the local ollama server is throwing a 500.

<!-- gh-comment-id:3508881785 --> @rick-github commented on GitHub (Nov 9, 2025): It appears this is an intermittent ongoing issue. The `invalid character 'I'` is a backend error that's causing `Internal Server Error` to be returned, and because it's not properly formed JSON, the local ollama server is throwing a 500.
GiteaMirror commented 2026-05-04 22:31:29 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Nov 10, 2025):

It should be fixed now, thanks @rick-github for all the help and everyone who reported! Let me know if you're still seeing the issue

<!-- gh-comment-id:3509181260 --> @jmorganca commented on GitHub (Nov 10, 2025): It should be fixed now, thanks @rick-github for all the help and everyone who reported! Let me know if you're still seeing the issue
GiteaMirror commented 2026-05-04 22:31:31 -05:00
Author
Owner
Copy Link

@kiril1976 commented on GitHub (Nov 11, 2025):

Error
500 Internal Server Error: unmarshal: invalid character 'I' looking for beginning of value. I still have this mistake and I dont know if this I something wrong with GUI or my MacBook OS Tahoe I use?

<!-- gh-comment-id:3516162338 --> @kiril1976 commented on GitHub (Nov 11, 2025): Error 500 Internal Server Error: unmarshal: invalid character 'I' looking for beginning of value. I still have this mistake and I dont know if this I something wrong with GUI or my MacBook OS Tahoe I use?
GiteaMirror commented 2026-05-04 22:31:33 -05:00
Author
Owner
Copy Link

@simmorsal commented on GitHub (Nov 11, 2025):

Getting this error when trying to upload an image using Open WebUI. Text works ok.

<!-- gh-comment-id:3516531758 --> @simmorsal commented on GitHub (Nov 11, 2025): Getting this error when trying to upload an image using Open WebUI. Text works ok.
GiteaMirror commented 2026-05-04 22:31:34 -05:00
Author
Owner
Copy Link

@rouk750 commented on GitHub (Nov 11, 2025):

I'm facing the same error with several cloud models; the latest failure occurred with gpt-oss:120b-cloud.
The same tests worked correctly with models running locally.

<!-- gh-comment-id:3516687613 --> @rouk750 commented on GitHub (Nov 11, 2025): I'm facing the same error with several cloud models; the latest failure occurred with gpt-oss:120b-cloud. The same tests worked correctly with models running locally.
GiteaMirror commented 2026-05-04 22:31:35 -05:00
Author
Owner
Copy Link

@kiril1976 commented on GitHub (Nov 11, 2025):

The problem is with Ollama GUI.I try all the models with Docker and Open WebUi and all the models works fine.So I will use them with Open WebUi.But they need to fix the Ollama GUi.

<!-- gh-comment-id:3516815714 --> @kiril1976 commented on GitHub (Nov 11, 2025): The problem is with Ollama GUI.I try all the models with Docker and Open WebUi and all the models works fine.So I will use them with Open WebUi.But they need to fix the Ollama GUi.
GiteaMirror commented 2026-05-04 22:31:37 -05:00
Author
Owner
Copy Link

@rouk750 commented on GitHub (Nov 11, 2025):

The problem is with Ollama GUI.I try all the models with Docker and Open WebUi and all the models works fine.So I will use them with Open WebUi.But they need to fix the Ollama GUi.
Not using GUI on my side. I am using langchain to consume model. I am exposing locally ollama through ollama serve

<!-- gh-comment-id:3516964370 --> @rouk750 commented on GitHub (Nov 11, 2025): > The problem is with Ollama GUI.I try all the models with Docker and Open WebUi and all the models works fine.So I will use them with Open WebUi.But they need to fix the Ollama GUi. Not using GUI on my side. I am using langchain to consume model. I am exposing locally ollama through ollama serve
GiteaMirror commented 2026-05-04 22:31:39 -05:00
Author
Owner
Copy Link

@xNeo-git commented on GitHub (Nov 11, 2025):

In my case, was solved for kimi-k2-thinking:cloud.
Same error with kimi-k2:1t-cloud.

Thanks!

<!-- gh-comment-id:3518461269 --> @xNeo-git commented on GitHub (Nov 11, 2025): In my case, was solved for kimi-k2-thinking:cloud. Same error with kimi-k2:1t-cloud. Thanks!
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#70678
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?