Meta's acquisition of Manus for $2.5 billion comes on the heels of Manus reporting impressive growth, with $100 million in annual recurring revenue (ARR) generated in just eight months and a staggering 147 trillion tokens processed since its launch. To understand the acquisition price, the article suggests looking at gross profit per token rather than traditional revenue metrics. In the AI sector, many companies operate with lower gross margins compared to typical software firms, which average around 71-72%. Manus's gross profit multiple, calculated at 50x, positions it favorably among peers. The comparison includes several AI inference companies, revealing insights into their valuations and operating efficiencies. For instance, DeepSeek, with 15 trillion tokens processed, has a valuation of $3.4 billion and a gross margin of 85%, resulting in a lower GP multiple of 20x. Conversely, Perplexity, which focuses on application-level monetization, commands the highest multiple at 222x despite a smaller token volume. Manus fits into this spectrum with its 50x multiple, reflecting a balance between token volume and gross margin efficiency. The analysis points out a strong correlation (0.70) between gross profit per token and company valuation, suggesting that investors prioritize monetization strategies over sheer token volume. However, the article notes that these findings are based on estimates and a limited sample size, indicating that while there is promise in this approach, it requires further validation. The figures and comparisons provided paint a clearer picture of the AI industry's evolving metrics for value assessment, especially as interest in companies like Manus grows.