Was really hoping this would be natively multimodal especially since its from open ai. but nope :/. At least the llama series does have something going for it