Indian agricultural warehousing and lending company Arya.ag is testing a system to tokenize warehouse receipts for stored grain on a dedicated Avalanche layer-1 blockchain, one of the largest attempts yet to bring physical commodity collateral onto distributed ledger infrastructure in India.
The company is working with Finternet to connect grain deposits, warehouse receipts, collateral commitments and loan status through the network. Devika Mittal, Ava Labs’ head of India, confirmed that testing is underway and said each tokenized receipt would represent ownership of the stored commodity. The companies did not disclose an expected launch date or how much grain or lending the initial deployment would cover.
Arya.ag stores about 2 billion USD in agricultural commodities across its warehouse network and supports approximately 120 billion Indian rupees — roughly 1.26 billion USD — in loans annually, according to the announcement. Its lending arm, Arya Dhan, issues about 230 million USD in loans each year. Those figures describe Arya.ag’s existing business and do not represent assets or loans already brought onchain.
Composite tokens for bank collateral
The mechanics center on what Finternet calls a composite token. Sanmesh Kalyanpur, a director at Finternet Labs, explained that Arya.ag’s samplers collect information about stored grain — quality, quantity, condition — and enter it into the company’s portal. Finternet will then combine farmer, commodity, warehouse and insurance information into a single token that banks can use when assessing collateral risk.
For lenders, the promise is a shared, tamper-resistant record showing what grain is stored, who owns it, whether it has already been pledged as collateral, and what debt is outstanding against it. Duplicate financing against the same warehouse receipt has long been a friction point in agricultural lending markets, and a unified onchain record is designed to make such double-pledging visible at a glance.
The announcement is candid about one limitation: the system will still depend on accurate verification of the physical commodities represented by the digital records. A blockchain can guarantee the integrity of the ledger, not the contents of the silo.
Finternet’s BIS pedigree
The Finternet concept is not a startup invention. It was outlined in a 2024 Bank for International Settlements working paper co-authored by Infosys co-founder Nandan Nilekani and then-BIS General Manager Agustín Carstens. The paper proposed interconnected unified ledgers for tokenized assets while emphasizing the need for supporting legal and regulatory frameworks — a framework-first philosophy that fits a use case where loan enforceability depends on courts recognizing digital receipts.
The choice of Avalanche is consistent with the chain’s growing real-world asset profile. The value of tokenized real-world assets on Avalanche exceeded 1.3 billion USD at the end of 2025, driven largely by loans and tokenized money-market funds, with BlackRock’s BUIDL fund among the contributors.
Why warehouse receipts matter
Electronic warehouse receipts allow farmers and agricultural businesses to borrow against commodities held in storage instead of selling them immediately after harvest — a practice that typically forces sales at seasonal price lows. Receipt-backed credit is one of the oldest forms of commodity finance, and digitizing it has been an Indian policy priority for years.
In 2024, the Indian government launched a 10 billion-rupee credit-guarantee program intended to encourage financing against electronic negotiable warehouse receipts, with a particular focus on small and marginal farmers. Tokenization is the logical next step of that arc: from paper receipts, to electronic receipts, to programmable receipts that carry their own collateral and insurance metadata.
The Arya.ag pilot joins a broader wave of traditional finance records moving onchain. Pineapple Financial recently placed 1 billion USD in mortgage records on Injective, and tokenized stocks tracked by RWA.xyz now exceed 2.9 billion USD in distributed value.
A cautious test, not a launch
Both companies are careful to frame the effort as a test. No grain volume, loan book size or go-live date has been committed, and the initial deployment’s scope remains undefined. What is notable is the stack being assembled: an established warehousing and lending operator, a blockchain infrastructure partner in Ava Labs, and a tokenization framework with direct lineage to BIS research.
If the composite token model works at Arya.ag’s scale — billions of dollars in stored commodities and over a billion dollars in annual lending — it would represent one of the most concrete demonstrations that blockchain-based collateral records can operate inside existing agricultural credit channels, rather than parallel to them.
For Avalanche, the deal extends a strategy of pursuing dedicated institutional deployments in Asia, following tokenized securities work in South Korea. For Indian agriculture, it offers a possible answer to a question that has hovered over farm credit for decades: how to make stored grain as liquid and as legible to banks as cash in an account.
2 billion USD in stored commodities and 120 billion rupees of annual lending behind this pilot. Even a small fraction of receipts going onchain would be a serious Avalanche use case
you should reread the article, it says those figures describe the existing business and nothing is onchain yet. testing = zero guarantees
Farmers will not care about Avalanche. They care whether Arya Dhan releases the loan faster when the warehouse receipt is tokenized. That turnaround time is the whole test.
Exactly. Arya Dhan moves 230 million USD in loans a year. If tokenized receipts cut disbursal from two weeks to two days, every competitor copies this within a season.
two weeks to two days on disbursal would genuinely change rural lending economics. the catch is whether warehouse staff can issue receipts onchain without opening new fraud vectors
tokenized warehouse receipts for grain, on a dedicated avalanche L1. if the composite token really bundles insurance and warehouse data, rural banks in india could finally lend against stored crops cheaply
no launch date, no volumes, no numbers on how much grain. testing underway is doing some heavy lifting here
fair, but even a pilot forces auditors to define receipt standards. the paper process has zero risk priced in because nobody ever measured it
collateral you can verify is the actual killer app and nobody wants to hear it because its not a dex
Finternet bundling the deposit, receipt, collateral commitment and loan status into one composite token is the unglamorous part that matters. Banks lend faster against paperwork they can audit instantly.
MSP connectivity in those warehouses is the part nobody asks about. the Avalanche L1 is fine, the hard problem is pushing receipts onchain from a yard with two bars of edge coverage.