m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA

ANITA-NEXT-24B-Magistral-2506-VISION-ITA is a 24 billion parameter Thinking Vision Language Model developed by m-polignano, built upon the ANITA family of LLMs. This model merges textual layers from ANITA-NEXT-24B-Magistral-2506-ITA with vision layers from Mistral-Small-3.1-24B-Instruct-2503, creating a Multilingual Vision Language Model (English + Italian). It is designed for further fine-tuning on specific Italian tasks, supporting a 32k token context length.