github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #5012] Seeded API request is returning inconsistent results #28931

New Issue
Closed
opened 2026-04-22 07:29:24 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 07:29:24 -05:00
Owner
Copy Link

Originally created by @ScreamingHawk on GitHub (Jun 13, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5012

What is the issue?

When using seed: 42069 and temperature: 0.0, I get consistent results only for the first ~126 characters. When using a seed, I would expect a deterministic result for the entire interaction.

    from ollama import Client, Options
    
    ollama_client = Client(host="http://localhost:11434")
    opts = Options()
    opts["seed"] = 42069
    opts["temperature"] = 0.0
    
    messages = [
        {"role": "user", "content": "Write a description of a cat"},
    ]
    
    res1 = ollama_client.chat(
        model="orca-mini",
        stream=False,
        messages=messages,
        options=opts
    )
    res2 = ollama_client.chat(
        model="orca-mini",
        stream=False,
        messages=messages,
        options=opts
    )
    
    # Fails
    assert res1["message"]["content"] == res2["message"]["content"]

OS

Windows

GPU

AMD

CPU

Intel

Ollama version

0.1.42

Originally created by @ScreamingHawk on GitHub (Jun 13, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5012 ### What is the issue? When using `seed: 42069` and `temperature: 0.0`, I get consistent results only for the first ~126 characters. When using a seed, I would expect a deterministic result for the entire interaction. ```py from ollama import Client, Options ollama_client = Client(host="http://localhost:11434") opts = Options() opts["seed"] = 42069 opts["temperature"] = 0.0 messages = [ {"role": "user", "content": "Write a description of a cat"}, ] res1 = ollama_client.chat( model="orca-mini", stream=False, messages=messages, options=opts ) res2 = ollama_client.chat( model="orca-mini", stream=False, messages=messages, options=opts ) # Fails assert res1["message"]["content"] == res2["message"]["content"] ``` ### OS Windows ### GPU AMD ### CPU Intel ### Ollama version 0.1.42
GiteaMirror added the bug label 2026-04-22 07:29:24 -05:00
GiteaMirror commented 2026-04-22 07:29:28 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Jun 14, 2024):

Hi there, sorry about this, I am looking into it. I'll merge with https://github.com/ollama/ollama/issues/4990

<!-- gh-comment-id:2168357167 --> @jmorganca commented on GitHub (Jun 14, 2024): Hi there, sorry about this, I am looking into it. I'll merge with https://github.com/ollama/ollama/issues/4990
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#28931
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?