DeepSeek's AI Model Generates Significant Revenue Per Hour

DeepSeek, a leading AI company, has revealed that its AI model generates $38 per hour for each graphics card used in its operations. This figure is significantly higher than the earnings from other AI models.

The company published this information as part of its inference engine release. This disclosure is notable because other laboratories typically do not share such financial details.

In comparison, a RTX 5090 graphics card, when rented out online, generates approximately $0.28 per hour for its owner. This means DeepSeek's earnings are roughly 135 times higher than the rental income from a single high-end graphics card.

DeepSeek's revelation highlights the potential profitability of AI models, particularly in terms of hardware utilization. The company's approach to transparency in financial performance is rare in the industry, where such details are often closely guarded.

This disclosure comes at a time when the AI industry is rapidly evolving, with companies continually seeking to optimize their models for better performance and cost-efficiency.

CONTEXT: DeepSeek's revelation of its AI model's earnings per graphics card is significant because it provides a rare insight into the financial viability of AI operations. This information can influence industry standards and practices, as other companies may seek to achieve similar levels of efficiency and profitability. Additionally, it underscores the growing importance of AI in various sectors, driving innovation and economic growth.