Orbio——An AI quota market emerging from Pons.

Written by: KarenZ, Foresight News

A fee generated from a token transaction can turn into an AI quota consumed for writing code and conducting research when it reaches Orbio.

According to Orbio's design, half of the ORBIO transaction fees will be used to support AI inference quotas. Once participants obtain quotas, they can either use them themselves or sell them at a discount to developers who need to invoke models. Thus, traders contribute transaction fees, quota holders activate idle balances, and AI users have the opportunity to reduce invocation costs.

This mechanism attempts to achieve an unusual transformation: turning the trading enthusiasm of the token market into AI resources that developers can genuinely use.

The market has begun to price this idea. As of September 21, 2026, GMGN data showed that the market capitalization of the ORBIO token was approximately $82 million, with today's market cap briefly rising to $90 million. For a project launched only about 20 days ago, the expectations from the market are already considerable.

The AI quota market emerging from Pons

To understand Orbio, one can momentarily set aside the tokens.

From a product perspective, Orbio is an AI quota market. Developers can purchase discounted quotas on the platform and then invoke different models through a unified interface. The official website listed 446 available models at the time, and after purchasing or obtaining quotas, users could use their Orbio API Key to invoke services.

Its service chain is also clear: users submit requests to the Orbio gateway, and Orbio utilizes its managed OpenRouter accounts to invoke upstream models. Therefore, Orbio's current business focus is on quota distribution, transaction, and usage entry, while the underlying model services still depend on OpenRouter and its connected providers.

On the cryptocurrency asset side, it comes from the Pons launching platform. The ORBIO token is issued through Pons on the Robinhood Chain, pairing the issuance market with tokenized NVDA.

As of September 21, by market capitalization, Orbio ranks second after the platform token PONS in the project list displayed by Pons on that day.

The background of the project's developers can also be traced. On the team side, the publicly participating builder of Orbio is Yash (X account @0x_aster). His personal GitHub profile lists development experience with the NFT perpetual contract DEX nftperp, and the official API documentation of nftperp (updated two years ago) directly mentions this account, corresponding to his previous development work.

On September 18, 2026, Jose (@The0xJose) announced his appointment as Orbio's adviser. His X personal profile states that he is the former founder of nftperp, co-founder, and product head of Pacifica. In a follow-up post to the announcement, Jose also mentioned that he had known Yash since the nftperp days and worked with him, and will continue to support his construction efforts.

Jose's assessment of Orbio primarily focuses on three directions: supporting Agents to continuously execute tasks on-chain with minimal human intervention, tokenizable and tradable inference quotas, and a market mechanism that can simultaneously connect quota suppliers and users. These views represent Jose's judgments upon joining the project and explain the problems Orbio aims to solve: how to allow AI quotas to be automatically acquired, circulated, and used by programs.

According to TrustMRR's project profile, Orbio was founded in September 2026, marked as a one-person team and self-funded. At least from the currently available public information, it remains a very small organization with rapidly iterating products in the early stages.

How does Orbio operate?

Orbio's mechanism has undergone a significant change. The early version mainly emphasized acquiring quotas by holding ORBIO tokens. On September 16, the project officially launched CREDIT, further clarifying the path of "staking ORBIO to gain CREDIT," and airdropped the previously accumulated usable inference balances to the corresponding wallets in the form of CREDIT.

The two tokens play different roles:

• ORBIO: Users can stake it to participate in earning CREDIT rewards.
• CREDIT: Priced based on each corresponding $1 worth of Orbio AI usage quota, it can be transferred, sold, or activated as API balance. Activation will destroy the corresponding token, and actual receipt must consider applicable protocol fees.

This means that those who want to use AI cheaply can purchase CREDIT directly without having to first buy or stake ORBIO. Token participants and AI users are connected through the quota market.

The source of discount funding is key to this design. According to the official documentation, half of the charged ORBIO transaction fees are used to support AI usage quotas, and stakers receive the corresponding CREDIT, which they can sell their unused portions.

CREDIT can be understood as a voucher for AI usage, where someone has already borne part of the cost. Sellers are willing to sell at a discount, while buyers wish to reduce model invocation expenses, and both parties settle transactions through the market.

However, the listed discount does not equate to the buyer's final savings ratio. The official website states that when purchasing quotas through the web retail, the platform will charge a 5% service fee based on the discounted price. Assuming a $100 quota is sold at $90, adding a $4.5 platform fee means the buyer pays a total of $94.5, saving 5.5% relative to the face value.

On September 19, Orbio announced a platform revenue reinvestment arrangement: half of the platform revenue is used for repurchasing and staking ORBIO, while the other half is used to purchase inference quotas and mint CREDIT based on that; the quotas acquired via the protocol will also be supplied to the order book at an 80% price.

This means that the Orbio platform itself will also become a quota supplier. Meanwhile, a link has been established between product revenue and ORBIO: after the platform earns revenue, it buys ORBIO, but the purchased tokens will be staked, rather than destroyed.

Another feature of CREDIT is that it facilitates automatic usage by software and Agents. Agents can purchase, receive, and activate CREDIT through smart contracts to supplement budgets for the next task without waiting for manual checkout. However, the on-chain activities are limited to the holding, trading, and activation of quotas; the actual model inference and balance accounting still rely on the Orbio gateway and its model suppliers.

How is Orbio performing?

Orbio has already publicly disclosed a set of data to observe the product's operational situation. As of September 21, 2026, 15:48, the official analysis page shows that the cumulative value of inference quotas generated by the platform is approximately $159,500, with about 249,500 requests served, processing about 20.5 billion model tokens, which include inputs and outputs of the models. Among these, the generated inference quotas reflect the value of the quotas, which cannot be regarded as platform revenue.

In terms of token participation and quota provision, approximately 354 million ORBIO are currently staked, accounting for 37.26% of the current total token supply. The high staking ratio indicates that many token holders are participating in CREDIT distribution; however, it does not prove that the quotas have been purchased or consumed by end users.

Data from the demand side needs further differentiation. The platform has recorded a total of 435 purchases and activations, which include both retail purchases and on-chain quota activations.

The comprehensive sales scale recorded on the page is approximately $13,150, but this data merges cash payments with on-chain activations calculated at face value. The official data clarification explicitly states that this indicator is not cash income.

Another observation window is TrustMRR. Its page shows that Orbio's cumulative income is $9,687, verified through the Whop API.

These data at least indicate that Orbio has seen real model invocation and quota circulation activities, and the product is not solely focused on token transactions. However, the project is still in a very early validation stage. To determine whether demand is robust, it is necessary to observe the number of independent paid buyers, repurchase rates, actual quota consumption amounts, and platform net income after deducting related costs.

What are the differences from Venice?

Venice is currently one of the more suitable projects for comparison with Orbio because both are trying to link AI usage rights with on-chain tokens, but the specific mechanisms are different.

According to Venice's current documentation, users can stake VVV to obtain sVVV, then lock sVVV to mint DIEM; staking one DIEM allows obtaining a daily usage quota of $1 for Venice. Quotas not used within an epoch do not roll over.

The differences in design are direct: CREDIT in Orbio resembles a transferable, sellable, and consumed pre-paid quota; whereas DIEM in Venice is more like an on-chain asset that continuously generates daily usage quotas. Therefore, while both are attempting to tokenize AI quotas, one cannot simply equate one CREDIT with one DIEM.

Venice has also accumulated more business experience. Its announcement on July 1, 2026, disclosed that there were 3.5 million registered users on the platform, processing approximately 1.3 trillion model tokens monthly, and developer API calls amounting to about 2 million times daily. These figures were official disclosures at that time and do not reflect real-time data as of September 21.

In terms of tokens, VVV currently has a market capitalization of $1.635 billion and an FDV of $2.77 billion. The gap in market capitalization between it and ORBIO cannot be directly translated into the latter's upside potential: the scales of both businesses, quota rights, supply structure, and value return methods are different.

For example, Venice announced in April a mechanism where new subscriptions that meet the criteria trigger VVV buybacks and destructions; Orbio's announcement on September 19 describes repurchases and staking. Destruction permanently reduces supply, while staked tokens still exist, so the effects on tokens cannot be conflated.

Conclusion

Orbio's attempt is to convert part of the transaction fees generated from token trades into usable and sellable AI quotas. Those who obtain quotas can use them themselves or sell them to developers needing to invoke models.

Compared to merely attaching an unimplemented AI concept to tokens, this mechanism is easier to validate: whether there are people in the market willing to purchase quotas, whether discounts can be sustained in the long-term, and whether acquired quotas are truly consumed will directly reflect product demand.

However, the current market pricing of ORBIO has clearly outpaced business data. What Orbio needs to prove next is whether users can consistently obtain stable price advantages, sufficient available quotas, and reliable invocation experiences.