TLDR: Chinese tech giant Baidu has announced it will open-source its flagship generative AI chatbot, Ernie 4.5, under an Apache 2.0 license. This strategic shift aims to accelerate AI adoption, expand its developer ecosystem, and challenge Western dominance by initiating a pricing war in the AI market. The move follows similar open-source initiatives by competitors like DeepSeek and positions Chinese labs as leaders in open AI development.
In a significant strategic pivot, Chinese technology behemoth Baidu has declared its intention to open-source its leading generative AI chatbot, Ernie 4.5. The move, which began with a gradual rollout around June 30, 2025, sees the model released under the permissive Apache 2.0 license, marking a departure from Baidu’s historical adherence to proprietary, closed AI systems.
This decision is poised to reshape the global AI landscape, intensifying pressure on AI model pricing and challenging the dominance of Western players such as OpenAI and Anthropic. Analysts, including Alec Strasmore of AI advisory firm Epic Loot Inc., have characterized Baidu’s action as ‘a declaration of war on pricing,’ signaling to startups worldwide that powerful AI access no longer necessitates premium costs. Baidu had already made Ernie free to use earlier in the year and subsequently slashed prices for its Ernie 4.5 and X1 models, with ‘Turbo’ variants seeing reductions of up to 80%.
The open-sourcing initiative is driven by multiple strategic objectives. Baidu aims to accelerate the adoption and innovation of its AI models by lowering access barriers for developers and researchers, thereby fostering a more collaborative open-source AI ecosystem. Robin Li, CEO of Baidu, articulated this goal plainly: ‘The goal is to let developers build powerful AI without high costs or vendor lock-in.’ This approach is also seen as a means to expand Baidu’s developer community, which currently lags behind competitors in API share.
Furthermore, Baidu’s move is interpreted as an attempt to capitalize on challenges faced by rivals. DeepSeek, a disruptor in the open-source AI space, has reportedly encountered delays in training its next-generation R2 model due to difficulties in acquiring sufficient Nvidia Corp. high-end graphics processing units, a consequence of U.S. export sanctions. By embracing open-source, Baidu may also seek to circumvent U.S. sanctions on China by leveraging the expertise of AI contributors worldwide, despite concerns from U.S. lawmakers who have labeled Chinese open-source AI efforts a ‘security risk.’
The Ernie 4.5 release includes a comprehensive model family, featuring ten versions with Mixture-of-Experts (MoE) architectures boasting up to 424 billion total parameters, alongside a compact 0.3 billion parameter model for lightweight tasks. The offering also includes ERNIEKit for development and FastDeploy for scalable deployment. Benchmarking indicates strong performance, with ERNIE-4.5-300B-A47B-Base outperforming Deepseek-V3-671B-A37B-Base on 22 out of 28 benchmarks. While direct comparisons to models like Claude 4 or OpenAI’s o3 are yet to be released, Baidu’s unreleased X1 model may be positioned for such evaluations.
Also Read:
- Google Unveils Gemini CLI: An Open-Source AI Powerhouse for Command Line Developers
- Microsoft Unveils Open-Source GitHub Copilot Chat for Enhanced AI Transparency in VS Code
This strategic shift by Baidu reflects a broader trend in China’s AI sector, where open-source models are gaining significant traction. Omdia analyst Lian Jye Su noted that Baidu’s decision was a ‘significant surprise’ given its history, adding that ‘disruptors like DeepSeek have proven that open-source models can be as competitive and reliable as proprietary ones.’ Tian Feng, president of Fast Think Institute, emphasized that ‘Through open source, Chinese AI models provide developers with an open platform for innovation and application, accelerating AI adoption and benefiting the entire tech ecosystem.’ This trend is poised to propel China’s AI sector forward rapidly, narrowing the technological gap with foreign peers and aligning with China’s broader strategy to build an autonomous and globally competitive AI ecosystem.
