DeepSeek V4-Flash Offers Major Cost Advantage Over Leading AI Models
DeepSeek challenges AI giants with dramatically cheaper model pricing.
Chinese artificial intelligence startup DeepSeek has introduced its latest AI model, V4-Flash, which research firm Artificial Analysis has described as one of the most cost-efficient well-known AI models available. The model has drawn attention for offering significantly lower operating costs compared with leading AI systems from companies such as OpenAI and Anthropic.
According to Artificial Analysis, DeepSeek V4-Flash costs around $0.14 per million input tokens and $0.28 per million output tokens. The research firm estimated that the model costs approximately 3 cents per benchmark test, making it considerably cheaper than competing systems including Moonshot AI’s Kimi K3, OpenAI’s GPT-5.6 Sol and Anthropic’s Claude Fable 5.
The benchmark comparison evaluates the overall cost of completing tasks by considering the amount of data processed and generated by AI models. Artificial Analysis said this provides a more practical measure than comparing only listed prices, as some models may require more computing resources to deliver responses despite having lower initial rates.
Also Read: OnePlus 16 Release Date, Expected Price, Features And Battery Details Leaked
DeepSeek V4-Flash received a score of 50 out of 100 on Artificial Analysis’ Intelligence Index, which evaluates AI systems across multiple areas including reasoning, coding and workplace-related tasks. The model performed at a similar level to Google’s Gemini 3.6 Flash and remained close behind other competitive models while maintaining a substantially lower operating cost.
The launch comes as DeepSeek attempts to strengthen its position in the rapidly expanding AI market. The company gained international attention in 2025 after its R1 model challenged assumptions about the scale of investment required to build competitive AI systems and contributed to market concerns over major technology companies’ AI spending.
DeepSeek is now facing growing competition from Chinese technology firms including ByteDance, Alibaba, Moonshot AI, MiniMax and Z.AI, as companies in China and the United States continue developing faster and more affordable AI models. Alibaba has also recently introduced Qwen3.8-Max, while DeepSeek is reportedly working on a more advanced V4-Pro model, though no launch timeline has been announced.
Also Read: Vivo S2 Teased Ahead Of Launch With Flagship-Level Battery And Performance Features
