DFDV: What DeFi Dev Corp's SEC Filings Show About Its SOL Treasury (2026)

DFDV is the Nasdaq ticker for DeFi Development Corp., a Boca Raton company that sold off a commercial real estate software business to buy Solana. As of 16 October 2026 it reported holding approximately 2,590,266 SOL and SOL equivalents. Every page ranking for the ticker is a price quote. None of them reads the filings, so this article does: every number below comes from the company's own SEC filings, from Solana mainnet, or from arithmetic on the two.

Two things are worth saying up front. First, DFDV publishes more about its treasury than almost any company in this category — weekly 8-Ks, defined per-share metrics, a dividend schedule down to the business day. Second, the interesting findings here come from putting those disclosures next to each other, not from anything hidden. This is analysis of public documents, not investment advice.

DeFi Development Corp. files with the SEC under CIK 0001805526, SIC code 6199 ("Finance Services"), and carries four tickers: DFDV (common stock, Nasdaq), DFDVW (warrants, Nasdaq), CHAD (Variable Rate Series C Perpetual Preferred Stock, Nasdaq) and DFUKF (OTC). The chief executive and chairman is Joseph Onorati.

The company's former names are on the same EDGAR record: it was Janover Inc. until April 2025 and Janover Ventures LLC before that. The pivot is documented in the Q2 2026 10-Q: on 31 March 2026 the board "approved the wind down of the legacy Janover Capital Markets and Janover Insurance businesses, which constituted substantially all of the operations of our Real Estate Platform segment." Substantially all of those operations ceased by the end of Q2 2026. On 26 June 2026 the company finished reincorporating from Delaware to Nevada, approved by written consent of holders of 81.79% of voting power.

What is left is a treasury vehicle. The 10-Q describes the strategy as "the acquisition, long-term holding, and active management of SOL and SOL-related digital assets," including "SOL, locked SOL, liquid staking tokens such as dfdvSOL, and other SOL-denominated or SOL-native positions," plus its own Solana validators. In H1 2026 the Real Estate Platform segment produced $320,000 of revenue against $5.66 million from the digital asset treasury.

There are two published figures and they measure different things.

The headline number comes from a weekly 8-K. The one filed 19 October 2026 reports 2,590,266 SOL and SOL equivalents as of 16 October. The balance-sheet number comes from the 10-Q for the quarter ended 30 June 2026, and it is much smaller:

Digital assets at fair value	30 Jun 2026	31 Dec 2025
SOL units (thousands)	888	1,093
Cost basis	$135,218K	$155,366K
Fair value	$65,156K	$136,000K

The gap is not a contradiction, and the company's own definition explains it. In the 5 October 2026 press release, DFDV defines net asset value as "the value of SOL and SOL equivalents, plus cash and cash equivalents, less SOL-denominated liabilities, outstanding principal of out-of-the-money convertible debt, other debt, and the aggregate notional amount of preferred equity." If SOL-denominated liabilities are subtracted at the NAV step, then the weekly treasury figure is a gross number reported before those liabilities — it counts SOL the company has pledged as collateral and SOL it has borrowed.

That matters because the balance sheet shows how much of the position is encumbered. At 30 June 2026 the company carried $101.6 million of "digital assets pledged as collateral" as a derivative asset against $89.8 million of digital asset financing arrangements as a derivative liability. The 10-Q also sets out the release schedule for locked SOL — SOL purchased below spot that cannot be withdrawn for a set period — at 337 thousand tokens remaining in 2026, 600 thousand in 2027 and 32 thousand in 2028, a schedule that "includes collateral related to digital asset financing arrangement term loans of 290.4 thousand tokens."

Worth noting for anyone comparing filings: "SOL and SOL equivalents" appears nowhere in the 10-Q, and the 5 October release defines SPS and NAV per share but not this term. It is an investor-relations measure, not a GAAP one.

Because the 8-Ks are weekly, the accumulation rate is a published time series. Stacking the last eight disclosures gives a shape the individual press releases do not:

As of	SOL and SOL equivalents	Added since prior
12 Aug 2026	2,311,523	—
27 Aug 2026	2,333,432	21,909
11 Sep 2026	2,388,923	55,491
18 Sep 2026	2,490,304	101,381
25 Sep 2026	2,538,010	47,706
2 Oct 2026	2,564,212	26,203
9 Oct 2026	2,586,282	22,070
16 Oct 2026	2,590,266	3,983

Weekly additions ran 101,381, then 47,706, then 26,203, then 22,070, then 3,983. Four consecutive decelerations, a 96% fall in the weekly add across five weeks, and the most recent 8-K marks the percentage change column "NM — amounts are not meaningful." The 21 September release described this as the "flywheel… continuing to gain momentum" and the chief executive said the company was "firing on all cylinders"; the 19 October filing reports the 3,983 week. Both statements are in the record, five weeks apart.

The timing lines up with the funding. The September surge coincided with the CHAD preferred stock offering, which was issued on 8 September. Once those proceeds were deployed, the add rate fell back toward what staking and validator rewards alone produce.

The fair-value table above is worth dividing out. At 30 June 2026 the 888 thousand SOL on that line had a cost basis of $135.2 million and a carrying value of $65.2 million. That is an average cost of about $152 per SOL against a mark of about $73 — the position was carried roughly 52% below what was paid for it. Note that the company buys locked SOL "below the spot rate," so the blended cost includes discounted purchases.

Two other accounting details are easy to miss and both favour reading the notes rather than the headline. First, purchases of Solana in H1 2026 were $0. The entire change in the fair-value line came from staking and validator rewards ($4.1 million), financing arrangements, collateral movements and mark-to-market. Second, there is a separate bucket of digital assets carried at cost less impairment rather than fair value, because liquid staking tokens sit outside the fair-value model:

At 30 Jun 2026	Gross	Impairments	Net	Fair value
dfdvSOL	$27,183K	$(11,520)K	$15,663K	$18,938K
apyUSD	$3,068K	—	$3,068K	$8,736K
Other	$9,520K	$(7,725)K	$1,795K	$4,500K

So part of the treasury is marked and part is impaired, and the impaired part is carried below its own disclosed fair value. Anyone building a model from the balance sheet needs both tables.

The equity line is the most direct read on what SOL's 2026 price action did to a leveraged holder. From the XBRL facts in the company's filings:

• Stockholders' equity: $243.8 million (30 Sep 2025) to $99.3 million (31 Dec 2025) to $10.2 million (31 Mar 2026) to negative $12.0 million (30 Jun 2026).
• Total assets: $203.3 million at 30 June 2026.
• Net loss: $110.7 million in H1 2026 ($27.3 million of it in Q2), after a $73.8 million loss for full-year 2025.
• Basic loss per share: $4.13 for H1 2026.
• Revenue: $5.98 million total for H1 2026, of which $5.66 million was digital asset treasury.

Against that $5.66 million of treasury revenue, H1 2026 interest expense was $4.30 million, plus $0.9 million of term-loan borrowing fees and $1.2 million of borrowing fees on DeFi financing arrangements. Treasury revenue did not cover financing costs in the first half, before a dollar of operating expense — and before CHAD, which did not exist until September.

This is the part no quote page shows. The capital stack, from the 10-Q:

Instrument	Outstanding (30 Jun 2026)	Rate
April 2030 convertible notes	$11.47M	2.5% coupon, 13.1% effective
July 2030 convertible notes	$114.58M	5.5% coupon, 6.4% effective
Digital asset term loans	$9.5M	12.5% on the open loan request
USDC open-term loan	$4.0M borrowed	10.0%
DeFi collateralized financing	$80.25M	Variable, set by each protocol
CHAD perpetual preferred	$15.81M stated	13%, rising to 14% in November

Three features of that stack deserve attention.

The SOL borrowings are expensive and margined. The 10-Q states the company borrowed SOL "with annual contractual borrowing fees ranging from 10.0% to 13.0%," posting collateral "with initial levels of 200.0% to 300.0% of the total loan value." For the term loan specifically: 12.5% annual fee, 250% initial collateral, 200% maintenance, and if coverage "declines to 150.0% or lower and is not remediated in a timely manner, the lender has the right to liquidate some or all of the posted collateral." Repaying early costs 20% of the borrowing fee that would have accrued to maturity.

The DeFi leg is recursive by design. In the company's words: "We may redeposit these borrowed assets into lending protocols and subsequently borrow additional assets against those deposits. This strategy is intended to increase our exposure to staking rewards, lending yields and other protocol incentives." The outstanding borrowings break down as SOL $56.2 million, PayPal USD $13.4 million, Global Dollar $7.5 million and other $3.2 million. If you have read our work on Kamino, marginfi or Save (formerly Solend), this is the same looping mechanic retail users run, at institutional size, inside a listed company.

The equity programmes have cost more than they raised. The equity line of credit with RK Capital Management is sized at up to $1.0 billion, expandable to $5.0 billion, with a $12.5 million commitment fee payable in stock. Through Q2 2026 the company had "recorded $6.3 million in commitment fees, or 1,585.2 thousand shares," and had issued 0.5 million shares under the facility "for approximately $1.9 million in net proceeds." The separate $200 million ATM sold 85,000 shares for $230,400 in Q2 and a further 393,000 shares for $1.1 million before the 10-Q was filed. Meanwhile the warrants issued with the April notes carry strikes of $17.14 and $21.43 and were all unexercised at 30 June.

One number cuts the other way, and it is a genuine positive: the company has been buying its own debt back cheaply. In Q2 2026 it paid $5.0 million to retire $7.9 million of July notes, booking a $2.8 million gain, and the August shareholder letter describes repurchasing about $3.5 million of principal for $2.3 million, "an approximately 35% discount to par."

CHAD is the most unusual thing in the filings. From the prospectus supplement dated 8 September 2026: 1,375,000 shares of Variable Rate Series C Perpetual Preferred Stock at a public offering price of $8.00, total $11.0 million, with "proceeds, before expenses, to us" of $7.72 per share. The underwriter, R.F. Lafferty & Co., later exercised its over-allotment for 206,250 more shares at about $1.65 million gross.

The key asymmetry is in the same document: the shares have a stated amount of $10.00, dividends "accumulate on the stated amount of the CHAD Stock, which is $10.00 per share," and the liquidation preference "will not be adjusted to an amount that is less than $10.00 per share." So dividends are computed on $10.00 while the company received $7.72.

Run the division. At the 14% rate effective in November, the annual dividend is $1.40 per share. Against the $8.00 offer price that is a 17.5% cost of capital; against the $7.72 the company actually received, 18.1%. The 15 October 8-K declares daily dividends of $0.00614 per share for 19 business days in November, "$184,469 in the aggregate for the full monthly period," which implies roughly 1.58 million CHAD shares outstanding — exactly the IPO plus the over-allotment, meaning the $300 million CHAD ATM had issued nothing as of that declaration. That is consistent with the company's stated intention that ATM issuance occur "at or above CHAD's $10.00 stated amount"; CHAD traded at $9.84 on 21 October 2026.

The marketing is itself a filed document. On 28 September 2026 the company filed a free writing prospectus containing the transcript of a radio advertisement: "Meet CHAD… currently offering a 13% annual dividend rate with dividends paid each business day when declared… because with CHAD, payday is every day." Other free writing prospectuses in the same series carry titles referencing a bank, a club, fast food and running.

And the company names the central risk itself, in its own forward-looking statements: "risks relating to CHAD, including that the dividend rate may be adjusted, that dividends are payable only when declared, that CHAD may trade below its stated amount… and that staking and validator rewards may be insufficient to support dividend payments."

That last risk factor is testable, because Solana publishes its own yield. Four unauthenticated calls to mainnet on 21 October 2026, at epoch 1064:

• Inflation rate: 3.592083% (getInflationRate).
• Total supply: 636,212,907.6 SOL (getSupply).
• Activated stake: 438,868,046 SOL across 665 current and 12 delinquent validators (getVoteAccounts).
• Staked share: 68.98%, so gross staking APR is 3.592% ÷ 0.6898 = 5.21% before validator commission.

That is the same derivation we walk through in our guide to Solana staking yields, and it sets the ceiling on what a SOL treasury earns from staking alone. Put the two sides together:

• Staking 2,590,266 SOL at 5.21% produces about 134,953 SOL a year — roughly $14.6 million at $107.88 per SOL.
• CHAD's annual dividend bill at 14% on 1.58 million shares is about $2.21 million, comfortably covered on that gross number, which is consistent with the company's statement that it "continues to maintain ample dividend coverage."
• But convertible coupons alone run about $6.6 million a year ($114.58M at 5.5% plus $11.47M at 2.5%), and SOL borrowed at 10.0% to 13.0% costs more than staked SOL earns at 5.21%.

The structure therefore cannot be funded by staking carry. The company does not claim otherwise — borrowed assets are deployed for "lending yields and other protocol incentives," not plain staking. But the arithmetic defines what has to go right: DeFi yields plus SOL price appreciation have to cover a cost of capital that runs from 2.5% at the cheapest layer to 18.1% at the newest one.

"DFDV mNAV" is one of the top related searches on the ticker, so here is the calculation with every input dated. On 21 October 2026, DFDV traded at $4.20 intraday (previous close $3.89; 52-week range $2.44 to $15.14) and SOL at $107.88.

• Market capitalisation: $4.20 × 30,997,790 shares (the 10-Q cover count, 13 August 2026) = $130.2 million.
• Gross treasury: 2,590,266 SOL × $107.88 = $279.4 million, or $9.01 of gross SOL per share. On that basis the market pays 47 cents per dollar of gross SOL.
• Claims against it: $126.05 million of convertible principal, $15.81 million of CHAD at stated amount, and $89.77 million of SOL-denominated financing liabilities (the most recent reported figure, 30 June 2026) = $231.6 million.
• Net: $279.4M − $231.6M = $47.8 million, or about $1.54 per share. Against a $130.2 million market cap that is roughly 2.7× net asset value.

Both figures are true, and the spread between 0.47× gross and 2.7× net is the liability stack. One caveat matters: the Q3 preliminary release reports "a reduction in notional SOL-denominated borrowings" and "more than 100% growth in NAV per share" since 12 August, so the June liability input is stale and the net figure is conservative. The Q3 10-Q will settle it. The method, though, is the point — any mNAV quoted without naming which liabilities it nets is not answering the question.

DFDV's internal scoreboard is SOL per share. The August 2026 shareholder letter reports SPS of 0.066 as of 12 August 2026, up about 24% year over year, and reaffirms "the Company's long-term target of 1.0 SPS by December 2028."

From 0.066 to 1.0 is a 15.2× increase in about 28 months, which requires compounding SOL per share at roughly 10% a month, every month, for two and a half years. Since SPS rises either from buying SOL faster than shares are issued or from staking rewards, and staking contributes about 5.21% a year, nearly all of that has to come from accretive issuance — selling stock above net asset value to buy SOL. That is exactly why the company talks about "establishing CHAD at its $10.00 stated amount": the ATM only works as described if the paper sells at or above par.

Two more things to track on the share count. The company repurchased 1,601,747 common shares for $10.5 million in H1 2026, and the 10-Q lists 18.9 million antidilutive shares — 6.1 million from convertible notes, 4.5 million warrants, 3.9 million dividend warrants, 3.8 million options and 0.6 million restricted stock units — against roughly 31.0 million shares outstanding. The company also discloses that SPS growth now applies an "updated, liability-adjusted methodology," and says plainly that SPS and NAV per share "do not represent GAAP book value per share, liquidation proceeds, or shareholder investment returns."

One claim in every DFDV press release can be checked directly: that it "operates its own validator infrastructure, generating staking rewards and fees from delegated stake." The liquid staking token is real and measurable. The dfdvSOL mint is sctmB7GPi5L2Q5G9tUSzXvhZ4YiDMEGcRov9KfArQpx, a Sanctum-issued token — the same infrastructure we cover in our Sanctum guide. On 21 October 2026 it had a supply of 1,849,570.84 dfdvSOL across 326 holders, and Sanctum's published SOL value of 1.076215 SOL per token implies about 1,990,500 SOL of underlying stake.

Be careful with that number: dfdvSOL is an open liquid staking token, so its pool includes third-party depositors and is not a measure of DFDV's own treasury. What it does show is a working LST of meaningful size, with a token-to-SOL ratio above 1.0 because rewards accrue into the value of each token rather than into the balance — the same design as mSOL. On the balance sheet, by contrast, DFDV's own dfdvSOL holding is carried at just $15.7 million net of $11.5 million of impairments. The validator business also appears in the intangibles note: validators acquired on 1 May 2025 at $3.6 million gross, amortised over three years, $2.2 million net at 30 June 2026.

Anyone searching the ticker is really asking how to get Solana exposure. The honest framing is that each route prices a different thing:

• DFDV common stock. Leveraged, operated SOL exposure inside a brokerage account, with a share count that changes, a convertible and preferred stack ahead of you, and a market price that has traded both well above and well below the SOL it represents.
• A spot Solana ETF. Unleveraged exposure with a disclosed fee, and a second staking fee layer we counted fund by fund in our Solana ETF analysis. No treasury strategy, no dilution, no mNAV.
• Holding and staking SOL yourself. The 5.21% gross figure above, minus validator commission, with no corporate layer at all — and the transaction costs are yours to manage.
• Trading it. Active SOL and memecoin trading on-chain, where the variable is execution rather than balance sheet. Our data on Solana copy trading outcomes is the clearest picture we have of how that goes in practice, and the Solana trading bot comparison covers the tooling.

A note on timing, since DFDV itself raised it: the 19 October press release framed its purchases around Alpenglow, Solana's consensus upgrade targeting roughly 150-millisecond finality, which the company said was scheduled to go live on 21 October 2026. Faster finality changes what on-chain execution can do; it does not change a cost of capital.

If what you actually want is Solana exposure you control, copying a verified on-chain trader with uwuu keeps the assets in your own non-custodial wallet, with sub-400ms execution and a performance-based fee — you pay only when you profit. No balance sheet, no preferred stock ahead of you, and no quarterly reconciliation of what "SOL equivalents" means.

All of it is free and reproducible:

• Filings index: data.sec.gov/submissions/CIK0001805526.json lists every DFDV filing. Send a User-Agent header or EDGAR refuses the request.
• The numbers: data.sec.gov/api/xbrl/companyfacts/CIK0001805526.json carries the tagged facts, including CryptoAssetNumberOfUnits, CryptoAssetFairValue, CryptoAssetCost and StockholdersEquity.
• The prose: the Q2 2026 10-Q is dfdv-20260630.htm under accession 0001805526-26-000055; the weekly treasury 8-Ks and their press-release exhibits sit in the same archive folders.
• The chain: getInflationRate, getSupply and getVoteAccounts against api.mainnet-beta.solana.com reproduce the 5.21% staking figure in four calls.
• The token: query the dfdvSOL mint for supply, and Sanctum's public SOL-value endpoint for the token-to-SOL ratio.

What does DFDV stock do?

DeFi Development Corp. (Nasdaq: DFDV) holds SOL as its principal treasury asset, stakes it, runs Solana validators and operates in DeFi. It was Janover Inc., a commercial real estate software business, until April 2025; the board approved winding down that legacy business on 31 March 2026, and in H1 2026 the real estate segment produced $320,000 of revenue against $5.66 million from the treasury.

How much SOL does DFDV hold?

Approximately 2,590,266 SOL and SOL equivalents as of 16 October 2026, per the 8-K filed 19 October. That is a gross figure: the company's own NAV definition subtracts SOL-denominated liabilities separately, and at 30 June 2026 the balance sheet showed $101.6 million of SOL pledged as collateral against $89.8 million of digital asset financing liabilities.

What is DFDV's mNAV?

It depends entirely on which liabilities you net. At $4.20 a share and SOL at $107.88 on 21 October 2026, the $130.2 million market cap is about 0.47× the $279.4 million gross treasury, but about 2.7× the roughly $47.8 million left after $126.1 million of convertible principal, $15.8 million of CHAD at stated amount and $89.8 million of SOL-denominated borrowings. Any mNAV quoted without naming its deductions is incomplete.

What is CHAD and how does its 14% dividend work?

CHAD is DFDV's Nasdaq-listed Variable Rate Series C Perpetual Preferred Stock, issued 8 September 2026. Dividends are cumulative, accrue daily on a $10.00 stated amount and are paid each business day when declared. The rate rises from 13% to 14% in November 2026, which is $1.40 per share a year. Because the shares were sold at $8.00 with $7.72 of proceeds to the company, that is a 17.5% to 18.1% cost of capital on the money raised.

Why is DFDV stock down from its highs?

The filings record what happened rather than why the market repriced it: a $110.7 million net loss in H1 2026, stockholders' equity falling from $243.8 million in September 2025 to negative $12.0 million at 30 June 2026, a SOL position carried about 52% below cost, and 18.9 million antidilutive shares outstanding against roughly 31.0 million shares. The stock's 52-week range is $2.44 to $15.14.

Is DFDV the same as a Solana ETF?

No. A spot Solana ETF holds SOL unleveraged and charges a disclosed fee plus, in most cases, a cut of staking rewards. DFDV is an operating company that borrows against its SOL, issues equity and preferred stock to buy more, and manages to a SOL-per-share target — currently 0.066, against a stated goal of 1.0 by December 2028.

Does DFDV really run its own validators?

Yes, and parts of it are checkable on-chain. Validators appear in the intangibles note at $3.6 million gross, acquired 1 May 2025 and amortised over three years. The dfdvSOL liquid staking token exists at mint sctmB7GPi5L2Q5G9tUSzXvhZ4YiDMEGcRov9KfArQpx with a supply of 1,849,570.84 tokens and about 1,990,500 SOL of underlying stake on 21 October 2026 — though as an open LST, that pool includes third-party depositors, not just the company's treasury.