[GH-ISSUE #7940] Mini-CPM-V-2.6-q8_0 produces incoherent responses after applying KV Cache q4_0 or q8_0. #67137

Open
opened 2026-05-04 09:32:22 -05:00 by GiteaMirror · 0 comments
Owner

Originally created by @SingularityMan on GitHub (Dec 5, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7940

What is the issue?

This happens when running ollama /generate via the python API. The output looks like the model is having a seizure. It seems to be able to see the images but its output is so random and erratic I can't make out anything from the text. I didn't change any other parameter about the model.

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.4.8 RC

Originally created by @SingularityMan on GitHub (Dec 5, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7940 ### What is the issue? This happens when running ollama `/generate` via the python API. The output looks like the model is having a seizure. It seems to be able to see the images but its output is so random and erratic I can't make out anything from the text. I didn't change any other parameter about the model. ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.4.8 RC
GiteaMirror added the bug label 2026-05-04 09:32:22 -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#67137