请问2.5-VL可以集成到ollama与open-webui中使用吗

[329724814]

No description provided.

[vladislavdonchev]

I am working on this, 7B is functional in llama.cpp, but I need some help with the 3B vision model extraction:
#869

For reference:
ollama/ollama#6564 (comment)
ggml-org/llama.cpp#11483 (comment)

After CLIP hardware acceleration is merged back into llama.cpp the changes can be propagated to ollama.
Ollama are also working on another multi-media inference engine, but there is no ETA on that as far as I know.