NewsFlash Articles Data Fundraising Skill&API

Base New Star POD: The Logic Behind a 14x Surge in One Month

Read this article in 14 Minutes

Dolphin's vision is to turn the world's idle GPUs into a "profitable AI computing power network."

Original Title: $POD: The Buyback Engine Powering the Dolphin Inference Network
Original Source: Dolphin
Original Translation: Yuliya, PANews

Editor's Note: Recently, the AI narrative of the Base ecosystem has seen explosive growth, with a focus on the privacy-first generative AI platform Venice ($VVV) and its ecosystem projects. As a co-developer of the default model in Venice Core, the Dolphin Network and its token $POD had an amazing performance in May, with the market cap skyrocketing from $12.2 million to $192 million, an increase of over 14x. This article details the Dolphin Network's unique "Peer-to-Pool" economic model, token value capture mechanism, and innovative design to ensure network security through staking and penalty mechanisms. The following is a detailed breakdown and analysis of the mechanisms:

Peer-to-Pool Economic Model Design

The Dolphin Network is designed as a "Peer-to-Pool" system, aiming to reuse everyone's idle GPUs. Each AI model runs on a GPU provided by the network.

This differs from most AI DePINs. In other networks, buyers usually directly rent a node from a provider, establishing a one-on-one "session."

· On the supply side, nodes running the same model form a "pool" to collectively process tasks sent by everyone. The system randomly assigns tasks based on the nodes' idle status, with no direct connection between task requesters and node providers. The sole criterion for nodes to earn rewards is how many AI computing tasks they have processed (inference tokens), with rewards paid in POD tokens from the protocol treasury.

· On the demand side, users utilizing the API can directly purchase capacity from the protocol. The Dolphin Network accepts various cryptocurrencies for payment, including $POD, $ETH, $BTC, $USDC, $XMR, and $ZEC.

All income received by the protocol will be used to buy back POD tokens on the market 100% — directly offsetting token issuance.

Buyers and sellers are separated, meaning that the POD rewards given to nodes may be more or less than the POD earned from income.

To provide a more intuitive illustration, let's consider a specific example of running the Qwen3.6-35B model on the Dolphin Network:

· Current cost of running datagen.dphn.ai: $0.50 per 1 million tokens processed.

· Cheapest comparable price on OpenRouter: $1.00 per 1 million tokens.

· Fee charged to users by Dolphin: $0.70.

· Fee paid to nodes by Dolphin: $0.50.

· Net buyback funds: $0.20 generated per 1 million tokens.

This means that the Dolphin Network's pricing is not only 30% lower than the cheapest centralized provider, but also generates $0.20 pure profit to buy POD in the market for every 1 million tokens generated.

Why is this the ideal use case for DePIN?

This model is seen as a highly promising application direction in the DePIN field, mainly for the following reasons:

· High AI Inference Demand: The market's thirst for AI inference power is currently at its peak.

· Massive Idle Compute Pool: There is an extremely large supply of idle gaming GPUs capable of running local AI models. This network model feels a lot like the old days of GPU mining (PoW), but because it produces AI computation with real commercial value, the money-making potential is much greater.

· Location Agnosticism: Unlike many DePIN networks, the geographical location of AI inference is not crucial, thus avoiding coverage issues. Due to the high flexibility of AI inference in terms of location, a delay of a few hundred milliseconds has minimal impact on user experience. This enables the Dolphin Network to connect consumers and computing resources globally, significantly enhancing the scalability and utilization of each node.

· The Inevitability of Liquidity Pooling Computing: This is the only way to unlock the largest group of GPU supply (gamers and PC enthusiasts). It allows nodes to come online or go offline at any time, without the need for fixed online times as required by P2P renting nodes. Previous GPU DePIN projects required a 1-to-1 binding between consumers and nodes, which is simply not feasible for idle GPUs like gaming PCs or data center GPUs because owners may want to take back their computers at any time. After all, no one wants to rent a GPU only to have the owner suddenly disconnect the virtual machine.

Tokenomics and Value Accrual

POD is the only valuable asset in the Dolphin ecosystem. All network revenue is automatically used to buy back POD in the market. Additionally, Dolphin has no shareholder-based equity structure and will never introduce one in the future.

For POD holders, staking the token in the xPOD Treasury provides multiple exclusive benefits:

· Direct automatic compound dividends from network token buybacks.

· Daily AI Inference allowance to freely use all models on the network.

· Premium subscription access in Dolphin's web chat, bots, and other ecosystem applications.

In the tokenomics design, Dolphin has drawn upon the essence of many excellent DeFi projects and deeply integrated those parts most suited to a distributed AI inference and training network:

· Learnings from the ETH Mechanism: Node operators and validators are required to stake a deposit, which will be slashed directly in case of malicious behavior (penalty).

· Learnings from the CRV Mechanism: Reward acceleration for node operators. Staking POD can double the earnings at most, and based on the deposit yield rates of other platforms, a 1.5 to 2x acceleration rate is highly competitive in the market.

· Learnings from the xSUSHI/yCRV Mechanism: Introduction of an automatically compounding staking treasury. Users do not need to manually claim rewards, meaning xPOD (the staked state of the Dolphin token) can directly act as collateral for node operators.

· Learnings from the stAAVE Mechanism: Reasonable withdrawal cooldown and drawdown time window are set to safeguard network fund stability.

· Learnings from the vlCVX/veCRV Mechanism: A "Bribe Market" is established for calculating the daily unused xPOD inference allowance. Users can sell their unused inference allowance to earn higher staking rewards.

Stake Locking, Violation Fine, and Reward Multiplier Mechanism

In a decentralized computing network, cheating is undoubtedly the biggest threat faced. If left unchecked, node operators may surreptitiously swap to smaller, crippled, or outright fake AI models and still claim rewards. As a result, output quality collapses, rented computational power users disappear, and the entire ecosystem's flywheel remains grounded.

To address this challenge, the Dolphin Network has introduced the "Stake Locking" mechanism, deeply linking node operators' vested interests with the value of the POD token. If malicious cheating is confirmed, the node will have a stake deducted equivalent to 4 weeks of income. This makes cheating economically unviable.

By default, node operators earn "Locked POD," which is in a staked state. Once a node has accumulated Locked POD equivalent to 4 weeks of income, they can choose during weekly settlements to either continue receiving Locked POD or receive liquid POD that can be freely traded.

If they choose to receive liquid POD, a 20% fee is deducted. This amount is directly deposited into the xPOD staking pool, distributed among other stakers and node operators who have dutifully locked their stake.

Nodes can further deposit xPOD into a bonding contract, which not only increases their earnings but also qualifies them for the validation of other nodes in the network.

The POD reward multiplier determines how much extra money a node can earn on top of the base reward. This mechanism was inspired by Curve Finance's Liquidity Provider (LP) acceleration mechanism, but Dolphin has made specific modifications for the decentralized AI network, incorporating features such as rewards based on usage, unified accounting of stakes across all accounts, and violation fines.

In summary:

· Nodes earn the base reward by completing AI computations, validation work, and related protocol tasks.

· The system multiplies the earned node reward by a factor based on the amount of staked tokens in your account and your earnings ratio.

· While calculating the earnings ratio, the system considers the average of your base rewards over the past few weeks and employs a "fast up, slow down" smoothing algorithm: When you take on more computing tasks, your average earnings metric rises rapidly; but when you are idle, it declines slowly.

· If your account holds a deposit of earnings for over 3 months and the active deposit is at least 50,000 POD, you are eligible to become a validator

· If your bonded deposit is equivalent to 6 months (26 weeks) of your earnings, the system guarantees that your reward will be at least multiplied by 1.5

· If your bonded deposit exceeds the earnings of 6 months, the maximum reward multiplier you can achieve is 2 times. The specific multiplier depends on the relative ratio between you and other over-bonders and the absolute amount you exceed the 6-month target

All calculations are based solely on the number of PODs, with no fiat price oracle involved in the reward system. Deposits are calculated per account (wallet), and the calculated reward multiplier applies to all nodes under your account name. If you add more nodes, your total account earnings will increase, so you need to proportionally increase your active deposit to maintain the original reward multiplier.

Finally, the Dolphin Network is set to release tomorrow a paper titled "Encrypted Live-Weight Proofs for Decentralized Inference." This paper will provide a detailed overview of a lightweight validation system that can verify the correct operation of various hardware-based nodes, surpassing the standard TEE validation only usable on enterprise NVIDIA GPUs.