The landscape of artificial intelligence is shifting as Chinese startup DeepSeek officially launches its V4-Flash model, positioning itself as the most budget-friendly option currently available on the market. According to research firm Artificial Analysis, the new model is remarkably cost-effective, with an average price tag of just 3 cents per test. This is a staggering efficiency improvement when stacked against competitors; for comparison, OpenAI’s GPT-5.6 Sol costs $1.86 to run, while Anthropic’s Claude Fable 5 demands $3.15. By focusing on the total computational cost—which accounts for how much data a model must process to generate a result—DeepSeek is making a compelling case for businesses that need to scale AI applications without breaking the bank.

Despite the focus on extreme affordability, the V4-Flash remains competitive in terms of capability. It achieved a 50 out of 100 on the Artificial Analysis Intelligence Index, placing it on par with Google’s Gemini 3.6 Flash. While higher-end models from OpenAI and Anthropic still lead in raw performance scores, DeepSeek’s strategy reflects the broader trend of a crowded Chinese market where startups like Moonshot and industry giants like Alibaba are aggressively fighting for dominance. As DeepSeek prepares to launch its more advanced V4-Pro, the industry is closely watching to see if this ultra-low-cost approach will successfully challenge the supremacy of major US AI firms in the global enterprise space.