[GH-ISSUE #10130] Ai Vision - Wildy Different Results Between Gemma3 34b OpenAI / Vs Ollama Endpoint #53160

Open
opened 2026-04-29 02:09:21 -05:00 by GiteaMirror · 0 comments
Owner

Originally created by @RamboRogers on GitHub (Apr 4, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10130

What is the issue?

I'm using the same inputs and prompts and am getting wildy different output from the Gemma3 model depending on the interface I'm using for Ollama.

OpenAI
Image

Ollama
Image

This is so weird. Its consistent as well, I re-run and re-run, same sort of theme with vision on boxes etc on the openai endpoint. The openai endpoint seems to be more accurate, while the ollama is more verbose.

Relevant log output


OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @RamboRogers on GitHub (Apr 4, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10130 ### What is the issue? I'm using the same inputs and prompts and am getting wildy different output from the Gemma3 model depending on the interface I'm using for Ollama. **OpenAI** ![Image](https://github.com/user-attachments/assets/d9ead990-5c1d-4767-aebb-c35a4f3c011a) **Ollama** ![Image](https://github.com/user-attachments/assets/fc13b17a-9a24-4ca9-a02a-aa7b93618dc2) This is so weird. Its consistent as well, I re-run and re-run, same sort of theme with vision on boxes etc on the openai endpoint. The openai endpoint seems to be more accurate, while the ollama is more verbose. ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-29 02:09:21 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#53160