github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-15 12:48:05 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #11535] qwen2.5vl:7b Not support restful /api/generate #54125

New Issue
Closed
opened 2026-04-29 05:15:10 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-29 05:15:10 -05:00
Owner
Copy Link

Originally created by @spwang84 on GitHub (Jul 26, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11535

What is the issue?

Hi team,

I found the restful API does not seem to work for the qwen2.5vl series

    ocr_data = {
        # "model": "gemma3:27b",
        "model": "qwen2.5vl:7b",
        "prompt": ocr_prompt,
        "images": [img_b64],
        "stream": False
    }

Always get the below result from any prompt, such as: identify the content or extract all words from the image.

<|im_start|>

Thank you.

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @spwang84 on GitHub (Jul 26, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11535 ### What is the issue? Hi team, I found the restful API does not seem to work for the qwen2.5vl series ocr_data = { # "model": "gemma3:27b", "model": "qwen2.5vl:7b", "prompt": ocr_prompt, "images": [img_b64], "stream": False } Always get the below result from any prompt, such as: identify the content or extract all words from the image. <|im_start|> Thank you. ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the macosbug labels 2026-04-29 05:15:11 -05:00
GiteaMirror commented 2026-04-29 05:15:12 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 27, 2025):

#!/usr/bin/env python3

import requests
import base64
import sys

ocr_prompt = "identify the content or extract all words from the image."

with open(sys.argv[1], "rb") as f:
    img_data = f.read()
    img_b64 = base64.b64encode(img_data).decode('utf-8')

ocr_data = {
    # "model": "gemma3:27b",
    "model": "qwen2.5vl:7b",
    "prompt": ocr_prompt,
    "images": [img_b64],
    "stream": False
}

response = requests.post("http://localhost:11434/api/generate", json=ocr_data)
print(response.text)

$ convert -size 300x100 xc:white -pointsize 20 -fill black -gravity center -annotate +0+0 "hello world" hello_world.jpg
$ ./11535.py hello_world.jpg | jq -r .response
The content of the image is:

```
hello world
```

Server logs may aid in debugging.

<!-- gh-comment-id:3124103467 --> @rick-github commented on GitHub (Jul 27, 2025): ```python #!/usr/bin/env python3 import requests import base64 import sys ocr_prompt = "identify the content or extract all words from the image." with open(sys.argv[1], "rb") as f: img_data = f.read() img_b64 = base64.b64encode(img_data).decode('utf-8') ocr_data = { # "model": "gemma3:27b", "model": "qwen2.5vl:7b", "prompt": ocr_prompt, "images": [img_b64], "stream": False } response = requests.post("http://localhost:11434/api/generate", json=ocr_data) print(response.text) ``` ````console $ convert -size 300x100 xc:white -pointsize 20 -fill black -gravity center -annotate +0+0 "hello world" hello_world.jpg $ ./11535.py hello_world.jpg | jq -r .response The content of the image is: ``` hello world ``` ```` [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may aid in debugging.
GiteaMirror commented 2026-04-29 05:15:13 -05:00
Author
Owner
Copy Link

@spwang84 commented on GitHub (Jul 28, 2025):

@rick-github

Very appreciate your reply.

I found the problem.

This issue only happens when the model qwen2.5vl series is running on Mac via Ollama, the NVIDIA GPU can work smoothly.

If I switch to other models, such as Gemma3 or Mistral, they are both OK on Mac.

I'm using Ollama 0.9.6 and did not try other Ollama versions.

<!-- gh-comment-id:3125231881 --> @spwang84 commented on GitHub (Jul 28, 2025): @rick-github Very appreciate your reply. I found the problem. This issue only happens when the model qwen2.5vl series is running on **Mac** via Ollama, the NVIDIA GPU can work smoothly. If I switch to other models, such as Gemma3 or Mistral, they are both OK on Mac. I'm using Ollama 0.9.6 and did not try other Ollama versions.
GiteaMirror commented 2026-04-29 05:15:13 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 28, 2025):

Server logs may aid in debugging.

<!-- gh-comment-id:3125237748 --> @rick-github commented on GitHub (Jul 28, 2025): [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 macos
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#54125
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?