Pixtral Large is a multimodal language model developed by French AI startup Mistral, which integrates a 123B text decoder and a 1B vision encoder. It has a context window of 128,000 tokens, allowing it to process large amounts of data in a single inference. Pixtral Large is built on the foundation of Mistral Large 2 and offers several key features such as a large context window, multi-resolution vision processing, unified evaluation protocols, instruction-tuned multimodal reasoning, seamless integration, scalability, and outperforms all open-models within its weight class on multimodal tasks. It has been evaluated on leading multimodal and text-only benchmarks, demonstrating competitive or superior results across tasks. Pixtral Large is available under two licenses: Mistral Research License for academic research and educational use, and Mistral Commercial License for commercial settings.