Ollama Support

[jordancaraballo]

Would be really nice to format the model to be supported as part of Ollama. Any plans for this?

[shuyansy]

Currently we don't have plans for Ollama support — TerraScope involves multimodal components (vision encoder + grounding/segmentation) that go beyond what Ollama's format natively supports.

We will release weights and code on Hugging Face soon. You can run inference directly through HF. Stay tuned!