Apertus has released Apertus 1.5, an updated version of its open-source AI model. This new iteration significantly expands its capabilities by incorporating multimodal understanding, allowing it to process images and audio alongside text. The model also features an enhanced 'thinking mode' for improved reasoning and a quadrupled context window of 262,144 tokens. Apertus 1.5 maintains its commitment to openness with publicly available weights, data, and training details. AI
IMPACT Enhances multimodal AI capabilities and pushes the boundaries of open-source model context windows, potentially accelerating research and development in these areas.
RANK_REASON This is a release of a new model version from a significant AI lab, Apertus, with notable improvements in capabilities. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
