QuixiAI/Devstral-Vision-Small-2507

Devstral-Vision-Small-2507, created by Eric Hartford at Quixi AI, is a 24 billion parameter multimodal language model with a 128k token context window. It combines the strong coding capabilities of Devstral-Small-2507 with the vision understanding of Mistral-Small-3.2-24B-Instruct-2506. This model is specifically designed for vision-augmented software engineering tasks, enabling developers to analyze UI screenshots and mockups to generate and debug code.