Centrifuge sets the comparison terms for a $14.2 billion market it holds roughly a ninth of

Centrifuge's total value locked sat between $1.5B and $1.7B in mid-to-late 2026 ^[12]. Against the $14.2B of onchain assets under management in the segment ^[3], that works out at 10.6% to 12.0% ^[2]. Total value locked and AUM are not the same measure, so read the range as an order of magnitude and not a market share. JTRSY, the Janus Henderson Anemoy Treasury Fund, crossed $1B in AUM at various points in 2026 ^[8], which is 59% to 67% of the platform's own locked value ^[3].

The comparison covers fee structures, redemption timelines, onchain liquidity and how well the tokens work inside DeFi protocols ^[2]. Some funds offer daily liquidity; others impose constraints that can lock capital for days ^[6]. Fees vary enough to eat into returns on instruments that are low-margin by design ^[7]. No fee levels or redemption windows for any named fund have been published ^[17], so the size of the dispersion the report is built on is still unknown.

Composability is where the products separate most. Pantera's assessment found only about 12% of tokenized assets scored highly enough to integrate smoothly with DeFi platforms ^[11], leaving 88% that did not ^[4]. That figure covers tokenized assets broadly, not treasury funds alone. It describes a class average and does not rank funds inside it.

There is also a limit that sits above fees entirely, and it comes from securities law. JPMorgan published an analysis in May 2026 finding that securities classification imposes transfer restrictions on tokenized money market funds that make seamless circulation more difficult than what stablecoins achieve ^[15]. Tokenized MMFs are classified as securities; in most jurisdictions stablecoins are not ^[14]. USDC goes to any wallet without permission, while a tokenized fund share runs into KYC gates, transfer agent requirements and jurisdictional barriers ^[16].

The operators name a different problem than the one an investor would. Centrifuge's own Tokenization Outlook 2026 surveyed 150 operators, and 86% said their primary challenge is scaling distribution, not creating new tokenized products ^[10]. If distribution is what binds, the marginal dollar of an operator's budget goes into integrations and sales. Centrifuge is spending on the first kind: a V3 multichain EVM architecture, and partnerships with Janus Henderson and Apollo ^[13].

In my view the fee-and-redemption lens is the right one for an allocator who might need the cash inside a week, because the yields cluster in a narrow range ^[5] and the assets underneath are the same short-term government paper, repo and T-bills ^[4]. For the segment's growth rate it has mattered less. Roughly $12.5B of new AUM arrived between June 2024 and mid-2026 ^[1] while about 88% of tokenized assets still failed the composability test ^[4]. If yields stop clustering, buyers go back to comparing yield and the wrapper recedes; if the transfer restrictions on fund shares are loosened, composability becomes an engineering question and the 12% moves.