Empty response on ollama

[hasaranga]

most of times responses are empty when using continuously with ollama. (fp16)

[vikhyat]

Would not recommend using the ollama version right now, it only supports a very old version of the model (from April). I need to reach out to them to figure out a path forward because I heard that llama.cpp might be dropping support for vision language models.

We have a Python client we're developing -- https://github.com/vikhyat/moondream/tree/main/clients/python -- only supports CPU for now but we're working on adding MPS and CUDA support as soon as possible./