Chinese tech giant Baidu has launched its latest open-source multimodal foundation model, ERNIE 4.5, and a reasoning model, ERNIE X1, which promises to deliver performance on par with DeepSeek R1 at only half the price. This move is expected to lower AI adoption barriers, intensify competition, and reshape pricing strategies across the industry, according to analysts.

The introduction of ERNIE 4.5 and ERNIE X1 "pushes the boundaries of multimodal and reasoning models," Baidu said in a release. ERNIE X1, in particular, is touted to deliver performance comparable to DeepSeek R1 at a significantly lower cost. This could be a game-changer for the AI industry, as cost-efficiency and accessibility become increasingly important considerations for developers and enterprises.

However, the success of these models will depend on several factors, including performance validation, developer adoption, and enterprise trust, according to Thomas Randall, research lead for AI at Info-Tech Research Group. "The global AI race is on, where cost-efficiency and accessibility become as important as raw capability," Randall noted.

Baidu plans to integrate both ERNIE models into its product ecosystem, including Baidu Search and other offerings. The company is also making its ERNIE Bot large language model (LLM) free to the public ahead of schedule, with enterprise users and developers able to access ERNIE 4.5 via APIs on Baidu AI Cloud's MaaS platform, Qianfan. ERNIE X1 is scheduled to be available on the platform soon.

Analysts are cautiously optimistic about the impact of ERNIE 4.5 and ERNIE X1. Jason Andersen, an analyst with Moor Insights & Strategy, noted that while the price-to-performance ratio of training the model is a consideration, it's unclear if that's enough to differentiate for developers or enterprises. "We will likely see AI hosting providers pick it up alongside other models. Once it's in the AWS or Azure toolsets, will it live up to the promise?" Andersen asked.

Randall believes that Baidu's open-sourcing of ERNIE 4.5 could accelerate innovation further via community-driven improvements, challenging the closed-source dominance of OpenAI and Google. This move also strengthens China's AI independence from US-based models, given the country's regulatory push for domestic AI dominance.

The launch of ERNIE 4.5 and ERNIE X1 is likely to intensify regional competition, with other Chinese tech giants like Alibaba, Tencent, and ByteDance expected to respond by investing further in this technology. Western firms may face new restrictions or competition in China, as businesses opt for homegrown models for compliance reasons.

Baidu highlighted the capabilities of ERNIE 4.5, which achieves improvements in "understanding, generation, reasoning, and memory, along with notable enhancements in hallucination prevention, logical reasoning, and coding capabilities." The model can integrate and understand text, images, audio, and video content, outperforming GPT-4.5 in multiple benchmarks while priced at just 1% of GPT-4.5.

The reasoning model ERNIE X1 supports various tools, including advanced search, Q&A on given documents, image understanding, AI image generation, code interpreting, webpage reading, TreeMind mapping, Baidu academic search, business information search, and franchise information search.

The launch of ERNIE 4.5 and ERNIE X1 comes just 10 days after Alibaba Cloud launched QwQ-32B, a compact reasoning model built on its latest large language model (LLM) Qwen2.5-32b, which delivers performance comparable to other large cutting-edge models, including those of Chinese rival DeepSeek and OpenAI's models, with only 32 billion parameters.