github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7446] MiniCPM-V 2.6 model crash with error code 500 when using ollama API in golang #51245

New Issue
Closed
opened 2026-04-28 19:02:00 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-28 19:02:00 -05:00
Owner
Copy Link

Originally created by @FreemanFeng on GitHub (Oct 31, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7446

What is the issue?

1730369280 debug

below is the golang code, set the prompt to be "请识别图片", load the attached image into []byte.

func RunVLM(prompt string, images ...[]byte) (bool, any) {
client, err := api.ClientFromEnvironment()
if err != nil {
log.Fatal(err)
}
model := "minicpm-v"

req := &api.GenerateRequest{
	Model:     model,
	Prompt:    prompt,
	KeepAlive: new(api.Duration),
	// set streaming to false
	Stream: new(bool),
}
for _, k := range images {
	s := base64.StdEncoding.EncodeToString(k)
	req.Images = append(req.Images, api.ImageData(s))
}
//req.KeepAlive.Duration = 24 * 60 * time.Minute

var v any
ctx := context.Background()
respFunc := func(resp api.GenerateResponse) error {
	// Only print the response here; GenerateResponse has a number of other
	// interesting fields you want to examine.
	fmt.Println(resp.Response)
	e := json.Unmarshal([]byte(resp.Response), &v)
	if e != nil {
		e = json.Unmarshal([]byte(fetchJSON(resp.Response)), &v)
		if e != nil {
			log.Println(e.Error())
			v = resp.Response
			return nil
		}
	}
	return nil
}

err = client.Generate(ctx, req, respFunc)
if err != nil {
	log.Println(err.Error())
	return false, nil
}
return true, v

}

When I run the code, ollama return "unmarshalling llm prediction response: invalid character 'e' looking for beginning of value" error.

So I debug the code and find that the ollama api service return 500 error.

OS

Windows

GPU

No response

CPU

Intel

Ollama version

0.3.14

Originally created by @FreemanFeng on GitHub (Oct 31, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7446 ### What is the issue? ![1730369280 debug](https://github.com/user-attachments/assets/24e7e4b7-cd2e-4453-885d-4e9e934fa8f3) below is the golang code, set the prompt to be "请识别图片", load the attached image into []byte. func RunVLM(prompt string, images ...[]byte) (bool, any) { client, err := api.ClientFromEnvironment() if err != nil { log.Fatal(err) } model := "minicpm-v" req := &api.GenerateRequest{ Model: model, Prompt: prompt, KeepAlive: new(api.Duration), // set streaming to false Stream: new(bool), } for _, k := range images { s := base64.StdEncoding.EncodeToString(k) req.Images = append(req.Images, api.ImageData(s)) } //req.KeepAlive.Duration = 24 * 60 * time.Minute var v any ctx := context.Background() respFunc := func(resp api.GenerateResponse) error { // Only print the response here; GenerateResponse has a number of other // interesting fields you want to examine. fmt.Println(resp.Response) e := json.Unmarshal([]byte(resp.Response), &v) if e != nil { e = json.Unmarshal([]byte(fetchJSON(resp.Response)), &v) if e != nil { log.Println(e.Error()) v = resp.Response return nil } } return nil } err = client.Generate(ctx, req, respFunc) if err != nil { log.Println(err.Error()) return false, nil } return true, v } When I run the code, ollama return "unmarshalling llm prediction response: invalid character 'e' looking for beginning of value" error. So I debug the code and find that the ollama api service return 500 error. ### OS Windows ### GPU _No response_ ### CPU Intel ### Ollama version 0.3.14
GiteaMirror added the bug label 2026-04-28 19:02:00 -05:00
GiteaMirror commented 2026-04-28 19:02:21 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 31, 2024):

Server logs will help in debugging.

<!-- gh-comment-id:2449548855 --> @rick-github commented on GitHub (Oct 31, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) will help in debugging.
GiteaMirror commented 2026-04-28 19:02:23 -05:00
Author
Owner
Copy Link

@jessegross commented on GitHub (Nov 5, 2024):

If you can try the 0.4.0 RC, we are now stricter about Unicode characters that are emitted so that will probably address this problem. LLMs can sometimes emit partial multi-byte Unicode characters.

<!-- gh-comment-id:2458279298 --> @jessegross commented on GitHub (Nov 5, 2024): If you can try the 0.4.0 RC, we are now stricter about Unicode characters that are emitted so that will probably address this problem. LLMs can sometimes emit partial multi-byte Unicode characters.
GiteaMirror commented 2026-04-28 19:02:24 -05:00
Author
Owner
Copy Link

@jessegross commented on GitHub (Nov 14, 2024):

I'm going to assume that this has been fixed by 0.4.0, please reopen if you still see it occurring on the current version.

<!-- gh-comment-id:2477399755 --> @jessegross commented on GitHub (Nov 14, 2024): I'm going to assume that this has been fixed by 0.4.0, please reopen if you still see it occurring on the current version.
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#51245
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?