github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-21 13:11:39 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #13389] Ollama Very Slow and Buggy 0.13.2 #86528

New Issue
Closed
opened 2026-05-10 03:35:27 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-10 03:35:27 -05:00
Owner
Copy Link

Originally created by @eliciel0513 on GitHub (Dec 9, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13389

What is the issue?

With version 0.13.2, overall stability has noticeably degraded and the system feels very buggy.

GPT-OSS 20B is experiencing severe issues such as repetitive looped thinking, frequent crashes, and unresponsive behavior.

qwen3-next:80b-a3b-thinking-q4_K_M is extremely slow and clearly not performing at the level expected when compared to larger models like GPT-OSS 120B. In practical terms, QWEN-3 Next takes roughly 5 minutes of thinking time at ~4 tokens/sec, whereas GPT-OSS 120B completes similar reasoning in about 1 minute at ~7 tokens/sec. This may be due to architectural differences, but the performance gap is significant.

Additionally, GPT-OSS 120B, which previously ran stably at 9–10 tokens/sec, is now consistently operating in the low 7 tokens/sec range under version 0.13.2.

Relevant log output

OS

Windows 11

GPU

RTX 3090

CPU

i9 14900k

Ollama version

0.13.2

Originally created by @eliciel0513 on GitHub (Dec 9, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13389 ### What is the issue? With version 0.13.2, overall stability has noticeably degraded and the system feels very buggy. GPT-OSS 20B is experiencing severe issues such as repetitive looped thinking, frequent crashes, and unresponsive behavior. qwen3-next:80b-a3b-thinking-q4_K_M is extremely slow and clearly not performing at the level expected when compared to larger models like GPT-OSS 120B. In practical terms, QWEN-3 Next takes roughly 5 minutes of thinking time at ~4 tokens/sec, whereas GPT-OSS 120B completes similar reasoning in about 1 minute at ~7 tokens/sec. This may be due to architectural differences, but the performance gap is significant. Additionally, GPT-OSS 120B, which previously ran stably at 9–10 tokens/sec, is now consistently operating in the low 7 tokens/sec range under version 0.13.2. ### Relevant log output ```shell ``` ### OS Windows 11 ### GPU RTX 3090 ### CPU i9 14900k ### Ollama version 0.13.2
GiteaMirror added the performancebugneeds more info labels 2026-05-10 03:35:28 -05:00
GiteaMirror commented 2026-05-10 03:35:33 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Dec 9, 2025):

Post the server log of a crash.

<!-- gh-comment-id:3632389101 --> @rick-github commented on GitHub (Dec 9, 2025): Post the [server log](https://docs.ollama.com/troubleshooting) of a crash.
GiteaMirror commented 2026-05-10 03:35:35 -05:00
Author
Owner
Copy Link

@mchiang0610 commented on GitHub (Dec 11, 2025):

@eliciel0513 I wanted to follow up on this. Is it possible to ask for more information on the server logs as @rick-github mentioned? It would help us in troubleshooting this. Thank you!

<!-- gh-comment-id:3643919328 --> @mchiang0610 commented on GitHub (Dec 11, 2025): @eliciel0513 I wanted to follow up on this. Is it possible to ask for more information on the server logs as @rick-github mentioned? It would help us in troubleshooting this. Thank you!
GiteaMirror commented 2026-05-10 03:35:36 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 14, 2026):

Closing as stale, please re-open with server logs if the problem persists.

<!-- gh-comment-id:3749144935 --> @rick-github commented on GitHub (Jan 14, 2026): Closing as stale, please re-open with server logs if the problem persists.
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 needs more info performance
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#86528
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?