DRVEXO-4B Unified is a new 4 billion parameter model that integrates image understanding, speech input and output, text conversation, and tool-use capabilities into a single package. It has demonstrated strong performance on various benchmarks, including a 77.93% score on a subset of MMLU-Pro. The model can process visual scenes, generate speech in Chinese and English, and make tool calls to query information or structure data according to JSON schema. AI
IMPACT This model's integration of multiple modalities and tool use in a compact 4B parameter size could enable more sophisticated and efficient AI applications.
RANK_REASON New model release from a lab. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →