[GH-ISSUE #10751] Output Discrepancy Between Ollama and llama.cpp for Same GGUF Model at Temperature=0.01 #7062

Closed
opened 2026-04-12 18:58:45 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @tangtang666 on GitHub (May 17, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10751

What is the issue?

I observe inconsistent outputs when running the same GGUF model , indicating that llama.cpp's results were correct. How should I find the reasons for these differences

Relevant log output


OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @tangtang666 on GitHub (May 17, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10751 ### What is the issue? I observe inconsistent outputs when running the same GGUF model , indicating that llama.cpp's results were correct. How should I find the reasons for these differences ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the needs more infobug labels 2026-04-12 18:58:45 -05:00
Author
Owner

@perfectecologietool commented on GitHub (May 17, 2025):

Try Set parameter temperature=0 and top_k=1

<!-- gh-comment-id:2888279368 --> @perfectecologietool commented on GitHub (May 17, 2025): Try Set parameter temperature=0 and top_k=1
Author
Owner

@rick-github commented on GitHub (May 17, 2025):

Some details would help to narrow the problem space. Model, configuration, inputs, outputs, logs.

<!-- gh-comment-id:2888402787 --> @rick-github commented on GitHub (May 17, 2025): Some details would help to narrow the problem space. Model, configuration, inputs, outputs, logs.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#7062