[GH-ISSUE #13449] Ollama 0.13.3 breaks abliterated Gemma-3 model #70935

Closed
opened 2026-05-04 23:30:13 -05:00 by GiteaMirror · 0 comments
Owner

Originally created by @twhlai on GitHub (Dec 12, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13449

What is the issue?

After I updated to 0.13.3, I get incoherent answers from the model hf.co/mradermacher/gemma-3-12b-it-norm-preserved-biprojected-abliterated-GGUF:Q5_K_M. It used to run fine in earlier versions.

server.log

Relevant log output


OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

0.13.3

Originally created by @twhlai on GitHub (Dec 12, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13449 ### What is the issue? After I updated to 0.13.3, I get incoherent answers from the model hf.co/mradermacher/gemma-3-12b-it-norm-preserved-biprojected-abliterated-GGUF:Q5_K_M. It used to run fine in earlier versions. [server.log](https://github.com/user-attachments/files/24137541/server.log) ### Relevant log output ```shell ``` ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version 0.13.3
GiteaMirror added the bug label 2026-05-04 23:30:13 -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#70935