mistral-small-3.2 OCR accuracy way too bad with llama.cpp compared to ollama?

Clarify: I tried the GGUF versions created by bartowski and others. Used these with llama.cpp and LM Studio. When you look at the list of files on Hugging Face, they all have a separate .mmproj file used for the vision functionality. Ollama on the other side embed this with their version of the model. Ollama gives some more details at https://ollama.com/blog/multimodal-models They say, "llama.cpp offers first-class support for text-only models. For multimodal systems, however, the text decoder and vision encoder are split into separate models and executed independently. Passing image embeddings from the vision model into the text model therefore demands model-specific logic in the orchestration layer that can break specific model implementations.Within Ollama, each model is fully self-contained and can expose its own projection layer, aligned with how that model was trained."