image-text-to-text

Unsloth releases a GGUF quantized version of the MiniMax-M3 multimodal model, enabling image-text-to-text tasks with support for Transformers, llama.cpp, vLLM, and other inference engines.