Mistral Unveils Trillion-Parameter Model While Google Slashes Image-Generation Costs in Tandem Releases

Mistral and Google unveiled major model updates on October 6, marking a shift in parameter scale and pricing strategy across the industry. Mistral released Mistral Large 4, nicknamed "Le Chonk," a 1-trillion-parameter multimodal model with 49 billion active parameters trained over two months on 4,000 Nvidia Grace Blackwell GPUs. The model is currently available via API at $1.36 per million input tokens, with open-weight release planned for October 27, 2026. Concurrently, Google rolled out Nano Banana 2.1, an image generation and editing model based on Gemini 3.6 Flash, across the Gemini app, Google AI Studio, Gemini API, Enterprise Agent Platform, and Search's AI Mode, pricing image outputs at approximately 50% less than its predecessor.