[GH-ISSUE #11] Add Ollama integration #8

Closed
opened 2026-10-26 21:09:53 +00:00 by bsakaguchi · 8 comments
Owner

Originally created by @vishalkrv on GitHub (Oct 19, 2024).
Original GitHub issue: https://github.com/deepseek-ai/Janus/issues/11

Can we get this on ollama, please ?

Originally created by @vishalkrv on GitHub (Oct 19, 2024). Original GitHub issue: https://github.com/deepseek-ai/Janus/issues/11 Can we get this on ollama, please ?
Author
Owner

@charlesCXK commented on GitHub (Nov 14, 2024):

Sorry for the late reply. We've been occupied with internal projects and haven't had time to migrate the model to Ollama yet. 😔

<!-- gh-comment-id:2477168747 --> @charlesCXK commented on GitHub (Nov 14, 2024): Sorry for the late reply. We've been occupied with internal projects and haven't had time to migrate the model to Ollama yet. 😔
Author
Owner

@limingchina commented on GitHub (Nov 15, 2024):

Is there still plan to support that?

<!-- gh-comment-id:2478101685 --> @limingchina commented on GitHub (Nov 15, 2024): Is there still plan to support that?
Author
Owner

@ozbillwang commented on GitHub (Jan 28, 2025):

could we re-open this request?

<!-- gh-comment-id:2617782530 --> @ozbillwang commented on GitHub (Jan 28, 2025): could we re-open this request?
Author
Owner

@agaldemas commented on GitHub (Jan 28, 2025):

Yep, I am also looking for a working janus model version for ollama ;O), the only one found is ozbillwang/deepseek-janus-pro, which give this error on my Mac M3:

ollama run ozbillwang/deepseek-janus-pro
pulling manifest 
pulling a338945d54b9... 100% ▕██████████████████████████████████████▏ 4.6 MB                         
pulling f0fd86e87432... 100% ▕██████████████████████████████████████▏  353 B                         
verifying sha256 digest 
writing manifest 
success 
Error: llama runner process has terminated: error loading model: missing tensor 'token_embd.weight

....so please to re-open this issue ! thanks in advance

<!-- gh-comment-id:2619618771 --> @agaldemas commented on GitHub (Jan 28, 2025): Yep, I am also looking for a working janus model version for ollama ;O), the only one found is ozbillwang/deepseek-janus-pro, which give this error on my Mac M3: ``` ollama run ozbillwang/deepseek-janus-pro pulling manifest pulling a338945d54b9... 100% ▕██████████████████████████████████████▏ 4.6 MB pulling f0fd86e87432... 100% ▕██████████████████████████████████████▏ 353 B verifying sha256 digest writing manifest success Error: llama runner process has terminated: error loading model: missing tensor 'token_embd.weight ``` ....so please to re-open this issue ! thanks in advance
Author
Owner

@ozbillwang commented on GitHub (Jan 29, 2025):

Yep, I am also looking for a working janus model version for ollama ;O), the only one found is ozbillwang/deepseek-janus-pro, which give this error on my Mac M3:

Please check if this works for you.

https://ollama.com/gguf/DeepSeek-Janus-Pro-7B

Image

<!-- gh-comment-id:2620393467 --> @ozbillwang commented on GitHub (Jan 29, 2025): > Yep, I am also looking for a working janus model version for ollama ;O), the only one found is ozbillwang/deepseek-janus-pro, which give this error on my Mac M3: Please check if this works for you. https://ollama.com/gguf/DeepSeek-Janus-Pro-7B ![Image](https://github.com/user-attachments/assets/88dd0fc5-6928-4710-9ae6-13aba3196c7d)
Author
Owner

@mTxBeN commented on GitHub (Jan 29, 2025):

Yes would love to see an ollama version of Janus Pro 7B please to run on local. Thanks 😊

<!-- gh-comment-id:2622154127 --> @mTxBeN commented on GitHub (Jan 29, 2025): Yes would love to see an ollama version of Janus Pro 7B please to run on local. Thanks 😊
Author
Owner

@ozbillwang commented on GitHub (Feb 6, 2025):

Yes would love to see an ollama version of Janus Pro 7B please to run on local. Thanks 😊

check this https://ollama.com/gguf/DeepSeek-Janus-Pro-7B

ollama run gguf/DeepSeek-Janus-Pro-7B

If compare to SD, Janus is loser. SD is faster.

<!-- gh-comment-id:2638671902 --> @ozbillwang commented on GitHub (Feb 6, 2025): > Yes would love to see an ollama version of Janus Pro 7B please to run on local. Thanks 😊 check this https://ollama.com/gguf/DeepSeek-Janus-Pro-7B ``` ollama run gguf/DeepSeek-Janus-Pro-7B ``` If compare to SD, Janus is loser. SD is faster.
Author
Owner

@agaldemas commented on GitHub (Feb 6, 2025):

Thanks, @ozbillwang, I maanged to load it and run (ollama run gguf/DeepSeek-Janus-Pro-7B) !
Nice but which UI tool can we use to generate images ?

Some links are welcome

<!-- gh-comment-id:2639116406 --> @agaldemas commented on GitHub (Feb 6, 2025): Thanks, @ozbillwang, I maanged to load it and run (ollama run gguf/DeepSeek-Janus-Pro-7B) ! Nice but which UI tool can we use to generate images ? Some links are welcome
Sign in to join this conversation.
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/Janus#8
No description provided.