iFlytek has announced the latest update to its deep inference model, "Spark X1." This update aims to improve its application capabilities in multiple fields, including mathematics, medicine, and education, incorporating numerous attractive new features and innovative improvements.
Spark X1's mathematical capabilities have been significantly enhanced, particularly in solving competition-level problems. Compared to other models, Spark X1 achieves comparable performance in mathematical tasks to DeepSeek R1 and related models from OpenAI, despite having fewer parameters. Users can directly verify its problem-solving abilities by inputting various mathematical problems through the iFlytek Spark website or app.
Simultaneously, the launch of the Spark Medical Large Language Model X1 marks a new stage in iFlytek's medical applications. Based on a comprehensive medical knowledge database and high-quality clinical data, the Spark Medical Large Language Model can provide more accurate advice on health consultations and disease diagnosis. By linking with users' health records, it enables more personalized responses, addressing the issue of impractical advice often found in traditional AI medical consultations.
Furthermore, the Spark Teacher Assistant and AI Judge Assistant have also been updated. The Spark Teacher Assistant has been optimized for lesson planning and classroom interaction, providing teachers with more logical and insightful guidance. Meanwhile, the AI Judge Assistant's legal reasoning capabilities have been strengthened, improving the accuracy of evidence review and judgment summaries, offering more reliable support for judicial work.
iFlytek, in collaboration with Huawei, also announced an upgraded version of the Spark integrated machine. This integrated machine not only supports dual-engine deep inference but also significantly improves inference performance, supporting application development across various industries.