[GH-ISSUE #2328] llava:34b is not working properly on my 36GB macbook M3 max #1345

Closed
opened 2026-04-12 11:10:55 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @microflyer on GitHub (Feb 2, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2328

When running the model with a picture, it returned ramdom text like below. I am using ollama version 0.1.22

/Users/danny/Downloads/ollama.png what is this
Added image '/Users/danny/Downloads/ollama.png'
username is first key ofthe.2!f+...0!5"0 g?..1...-
. 10dd.. t1.!.
....
...!:/.-.s[..,.,.:..)
. A.---..... .-!:
(^C

I have tried the other models, yi:34b, llava:13b, mistral... all the others are working perfectly.

Originally created by @microflyer on GitHub (Feb 2, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2328 When running the model with a picture, it returned ramdom text like below. I am using ollama version 0.1.22 >>> /Users/danny/Downloads/ollama.png what is this Added image '/Users/danny/Downloads/ollama.png' username is first key ofthe.2!f+...0!5"0 g?..1...- . 10dd.. t1.!. .... ...!:/.-.s[..,.,.:..) . A.---..... .-!: (^C I have tried the other models, yi:34b, llava:13b, mistral... all the others are working perfectly.
Author
Owner

@mxyng commented on GitHub (Feb 2, 2024):

There's a bug that impacts llava v1.6. It'll be fixed in the next release (coming very soon). See #2296

<!-- gh-comment-id:1924254254 --> @mxyng commented on GitHub (Feb 2, 2024): There's a bug that impacts llava v1.6. It'll be fixed in the next release (coming very soon). See #2296
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#1345