Recently, discussions about the Walrus project have heated up in the institutional circle, not only because of Grayscale's mention but more importantly because of the interesting economic model design behind it.

First, let's talk about Grayscale's actions. In July this year, Grayscale included Walrus in its watchlist, directly calling it the "Sui data layer core." On the day the news was announced, WAL's market cap increased by 40%. This reflects not short-term speculation but the recognition by large funds of the long-term value of decentralized storage as a track.

But what truly deserves attention is WAL's token design. Let's look at the total supply—5 billion. At first glance, that's quite a lot, but there's a key mechanism: every time someone stores data, the system destroys a portion of tokens. In other words, the higher the network usage frequency, the less circulating supply there is. This deflationary logic is somewhat similar to Ethereum's EIP-1559 burn mechanism but directly linked to storage activity. More data uploads make the token scarcer, creating a positive feedback loop of demand and scarcity.

Now, how is the pricing determined? Walrus adopts a market bidding model. After nodes submit their bids, the system calculates the median weighted by stake. For example, if six nodes bid $0.5, $0.9, $1.0, the final price is set at $1.0. This approach protects node revenue expectations and prevents malicious low-price competition, maintaining market order while allowing market dynamics.

A truly overlooked highlight is the subsidy leverage. Walrus allocates 10% of tokens for storage subsidies, enabling developers to migrate data at extremely low costs. Imagine—NFT platforms subsidizing storage costs nearly to zero, while users participate in network fee sharing through staking WAL. Once this closed loop kicks in, a positive cycle of "subsidy → ecosystem expansion → increased token demand" forms. This logic was also used early on by Filecoin, but Walrus benefits from Sui's high TPS, solving Filecoin's old problem of slow transaction confirmation.

Of course, risks are also present. Circulating supply accounts for only 29.57% of the total supply, and subsequent token unlocks could bring selling pressure. Therefore, what truly matters is not the short-term K-line but the actual activity on the mainnet—whether storage transaction volume is high and whether the number of nodes is growing. These are the real indicators of the project's health.