[GH-ISSUE #4234] Customized LLaVA Setup #2642

Closed
opened 2026-04-12 12:59:18 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @zhangry868 on GitHub (May 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4234

I wonder whether there is a guideline on hosting customized LLaVA model. I have both mmprojector and base models gguf files. Feel free to point me any related materials/links.

Many thanks,
Rui

Originally created by @zhangry868 on GitHub (May 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4234 I wonder whether there is a guideline on hosting customized LLaVA model. I have both mmprojector and base models gguf files. Feel free to point me any related materials/links. Many thanks, Rui
GiteaMirror added the feature request label 2026-04-12 12:59:18 -05:00
Author
Owner

@jmorganca commented on GitHub (May 7, 2024):

Hi Rui,

For using LLaVA models with ollama create, check out this example: https://huggingface.co/xtuner/llava-phi-3-mini-gguf/blob/main/OLLAMA_MODELFILE_F16 which shows how to build an Ollama model from both the language and projector files.

Hope this helps!

<!-- gh-comment-id:2099492719 --> @jmorganca commented on GitHub (May 7, 2024): Hi Rui, For using LLaVA models with `ollama create`, check out this example: https://huggingface.co/xtuner/llava-phi-3-mini-gguf/blob/main/OLLAMA_MODELFILE_F16 which shows how to build an Ollama model from both the language and projector files. Hope this helps!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#2642