[GH-ISSUE #4639] Prompt caching causes reproducible outputs to be inconsistent #28675

Closed
opened 2026-04-22 07:10:13 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @jmorganca on GitHub (May 25, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4639

Originally assigned to: @jmorganca on GitHub.

What is the issue?

When trying to generate reproducible outputs, changing prompts causes results to be inconsistent

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @jmorganca on GitHub (May 25, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4639 Originally assigned to: @jmorganca on GitHub. ### What is the issue? When trying to generate [reproducible outputs](https://github.com/ollama/ollama/blob/main/docs/api.md#request-reproducible-outputs), changing prompts causes results to be inconsistent ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-22 07:10:13 -05:00
Author
Owner

@jmorganca commented on GitHub (Jun 11, 2024):

Fixed by https://github.com/ollama/ollama/pull/4986

<!-- gh-comment-id:2161628506 --> @jmorganca commented on GitHub (Jun 11, 2024): Fixed by https://github.com/ollama/ollama/pull/4986
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#28675