China’s tech giant Baidu has launched two new artificial intelligence (AI) models, including a reasoning-focused model designed to rival DeepSeek’s offerings, as it seeks to solidify its position in the rapidly evolving AI landscape.
The new model, ERNIE X1, is touted as a powerful deep-thinking AI capable of autonomous tool usage. Baidu claims that ERNIE X1 delivers performance comparable to DeepSeek R1 but at half the price. The company highlights the model’s ability to excel in understanding, planning, reflection, and evolution, making it a significant advancement in reasoning-based AI development.
Alongside ERNIE X1, Baidu also introduced ERNIE 4.5, an upgraded version of its foundational AI model. According to Baidu, ERNIE 4.5 features enhanced multimodal understanding, meaning it can process and integrate diverse data formats, including text, video, images, and audio. The model reportedly boasts improved capabilities in language processing, logic, memory, and content generation. Additionally, Baidu claims ERNIE 4.5 possesses a high emotional intelligence (EQ), allowing it to understand network memes and satirical cartoons a feature that could improve human-AI interactions.
Baidu was among the first Chinese tech companies to launch a ChatGPT-style chatbot, yet it has faced challenges in gaining widespread adoption for its Ernie large language model (LLM). Despite asserting that its AI performance is comparable to OpenAI’s GPT-4, Baidu has struggled to outshine its competitors in China’s fiercely competitive AI sector.
One of Baidu’s biggest challenges comes from DeepSeek, a Chinese AI startup that has shaken the industry by rolling out high-performing models at a fraction of the cost of their Western counterparts. DeepSeek’s aggressive expansion has re-energized the global AI race, pushing major tech firms, including Baidu, to continuously innovate.
The introduction of ERNIE X1 and ERNIE 4.5 underscores Baidu’s ambition to lead China’s AI sector and compete with both domestic and international AI giants. As the race for more advanced, cost-effective AI models heats up, companies like Baidu will need to ensure their models not only perform well but also achieve greater adoption and practical applications in real-world scenarios.
