# 128k long context processing

Granite-4.0-Tiny-Base-Preview is a 7-billion parameter Mixture of Experts (MoE) language model developed by IBM, featuring a 128k token context window and enhanced expressive capabilities through Mamba-2 technology.

The Gemma 3 27B IT model, introduced by Google, is suitable for various text generation and image understanding tasks, supporting a context length of 128k tokens and multimodal image processing.

An enhanced version of QwQ-32B, improving reasoning and output capabilities through 'cubed' and 'triple-cubed' methods, supporting 128k context.

Mistral-Nemo-Instruct-2407 is a large language model fine-tuned on instructions based on Mistral-Nemo-Base-2407, jointly trained by Mistral AI and NVIDIA, outperforming existing models of similar or smaller size.