mistralai/Mistral-Nemo-Instruct-2407

Mistral-Nemo-Instruct-2407 is a 12 billion parameter instruction-tuned large language model developed jointly by Mistral AI and NVIDIA. It is based on the Mistral-Nemo-Base-2407 architecture, featuring a 32768 token context window and Grouped-Query Attention (GQA). This model is optimized for instruction following and excels across various benchmarks, including MMLU, HellaSwag, and multilingual tasks, making it suitable for general-purpose conversational AI and code-related applications.