This post uses hypothetical scenarios for illustrative purposes only. It does not describe any actual client, transaction, or representation, and is not legal advice.
Download: SAFT vs T-SAFE vs Token Warrant — Comparison Worksheet (.docx) — a companion resource for this post. Adapt with counsel before use.
A common 2026 crypto raise pattern looks like this. A pre-launch protocol team, roughly a year from mainnet, has strong seed traction and a term sheet on the table. The lead wants tokens, not just equity. The founders want to preserve optionality — a foundation may or may not be spun up, the tokenomics still have three open drafts, and the SEC’s posture is drifting month to month. The lawyers in the room reach for the same three instruments they have been reaching for since roughly 2018 — the SAFT, the SAFE-plus-token-warrant, and the standalone token warrant — but with much sharper edges after LBRY, Kik, Telegram, Ripple, and Terraform. Picking the wrong one at the seed stage compounds into serious problems by Series B and TGE.
Here is how the three instruments actually differ, when each is the right tool, and what has changed after five years of enforcement.
First — the SAFT
The Simple Agreement for Future Tokens was drafted in 2017 as an attempt to thread a needle. The company would sell a security to accredited investors under Reg D. That security would represent a right to receive tokens once the network was launched. At the moment of network launch, the tokens themselves — the theory ran — would be sufficiently decentralized that they would no longer be securities in the hands of holders. Investors got early token exposure; the company got fundraising under a familiar exemption; and everyone got to sidestep the ugly question of whether the tokens themselves were securities.
The theory did not survive contact with the SEC. In SEC v. Kik Interactive, Inc., No. 19-cv-5244 (S.D.N.Y. 2020), Judge Hellerstein held that the SAFT and the underlying token sale were parts of a single integrated offering — a security. In SEC v. Telegram Group Inc., No. 19-cv-9439 (S.D.N.Y. 2020), Judge Castel granted a preliminary injunction on the same integrated-offering rationale. And in SEC v. LBRY, Inc., No. 21-cv-260 (D.N.H. 2022), the court held on summary judgment that the token itself was a security under SEC v. W.J. Howey Co., 328 U.S. 293 (1946), regardless of the labeling. The SAFT structure did not immunize the token; it merely delayed the securities-law reckoning.
What survived is a much narrower use case. The SAFT still makes sense when the raise is genuinely token-only — no priced equity, no equity kicker — and the protocol will launch within a reasonable horizon. It is documented under Reg D 506(c) or Reg S, treats the token as a security on issuance, and requires that the company plan its post-launch compliance program (transfer restrictions, KYC, geo-fencing) as if the token remains a security in secondary markets. Post-Ripple, that assumption is safer than the alternative. In SEC v. Ripple Labs, Inc., No. 20-cv-10832 (S.D.N.Y. 2023), Judge Torres drew a line between institutional sales (which were securities) and programmatic sales on exchanges (which were not, on that record) — a distinction later criticized in SEC v. Terraform Labs Pte. Ltd., No. 23-cv-1346 (S.D.N.Y. 2023), and rendered less load-bearing by the SEC’s evolving 2025–2026 posture on crypto asset securities. Founders using a SAFT today should not rely on the programmatic-sale carveout in their diligence memos.
Second — the SAFE plus token warrant (the “T-SAFE”)
The most common seed instrument for high-growth crypto companies in 2026 is not the SAFT but the SAFE paired with a token warrant — informally, the T-SAFE. The mechanics are straightforward. The investor writes a check on a standard Y Combinator or post-money SAFE, converting into preferred equity on the next priced round. Attached to the SAFE is a separate warrant granting the investor the right to receive a pro rata share of the network’s token supply at TGE, typically capped at the investor’s percentage of fully-diluted equity, and often with a nominal exercise price.
The T-SAFE is the belt-and-suspenders combo. If the company never launches a token, the investor still owns equity. If the company launches a token and the equity ends up subordinated to the token’s economic value, the investor participates in the token upside. The two instruments are legally separate — one is convertible equity, the other is a contract right for future tokens — but they are usually delivered together, referenced in the same side letter, and diligenced together in later rounds.
The Howey analysis is different for each piece. The SAFE is a security under standard convertible-equity precedent. The warrant, at grant, is a contract right — not a token, and not yet a security in the SEC’s traditional framing, though the exercise of the warrant at TGE triggers a fresh securities analysis for the tokens delivered. The company usually documents the warrant exercise as either (a) an exempt distribution to accredited holders under Reg D, or (b) a distribution by the foundation rather than the operating company, if a foundation has been formed. The choice affects who bears the securities-law risk and where indemnities are drafted.
Two subtler points matter. First, if the token warrant is issued by the operating company but the tokens will be issued by a foundation, the warrant needs an assignment or novation mechanic — otherwise the operating company cannot deliver what the warrant obligates it to deliver. Second, the pro-rata percentage in the warrant is usually calculated against fully-diluted supply at TGE, not fully-diluted equity — a distinction that matters when the token cap table and the equity cap table drift apart, as they always do.
Third — the standalone token warrant
The standalone token warrant is what shows up attached to a priced equity round — Series Seed, Series A, sometimes Series B. The company is raising priced equity for cash and giving each investor, alongside the preferred shares, a warrant to purchase tokens at TGE at either a discount to the public TGE price or a fixed strike. The warrant is negotiated as an addendum to the equity round rather than as a fundraising instrument in its own right.
This is the cleanest structure of the three for later diligence. There is a familiar priced equity round with familiar documents. The warrant is a discrete addendum. The Howey posture is the equity posture — Reg D 506(b) or 506(c) at the round — plus a separate token-side analysis at exercise. Series B investors and M&A diligence teams do not have to reconstruct the theory of an integrated pre-launch token sale, because there was none.
The tradeoff is that the standalone warrant only works when there is a priced round to attach it to. Pre-priced-round, before there is a Series Seed or a Series A, the SAFT or the T-SAFE are the only serious options.
Vesting, lockups, and transfer restrictions
All three instruments interact with the same downstream mechanics. On the founder side, the standard 2026 posture is a four-year monthly vest with a one-year cliff, paired with a post-TGE lockup of six to twelve months for the founders and often longer for the team. See our companion post on founder token vesting for the drafting mechanics. On the investor side, lockups are typically shorter — six months from TGE is common — but the specific curve is negotiable in the warrant and the SAFT.
Transfer restrictions matter more than founders often expect. If the token is delivered under Reg D, it carries the same one-year holding-period restriction as any other Reg D security under Rule 144. If it is delivered under Reg S to non-U.S. persons, it is subject to a distribution-compliance period. If a company plans to list on a centralized U.S. exchange promptly after TGE, the interaction with these restrictions has to be modeled — otherwise the market maker gets tokens it cannot legally trade in the first months of the market.
Interaction with founder equity
The instrument choice ripples backward into the founder cap table. Under a SAFT-only raise, the company is diluting only the token — the equity cap table is unchanged, but the founders’ economic exposure is now split between equity (which may never see meaningful value if the token accrues most of the network’s economics) and tokens (subject to their own vesting, lockup, and tax posture). Under a SAFE plus warrant, both cap tables move — equity dilutes on SAFE conversion, tokens dilute on warrant exercise. Under a standalone warrant attached to a priced round, both cap tables move simultaneously, and the founders can plan around a coordinated dilution schedule.
The tax posture is a separate topic — IRC § 83(b) elections behave differently for token grants than for equity, because the token often does not exist at grant and has no ascertainable value. That is worth its own memo and is beyond the scope of this post.
The SEC’s evolving stance
The SEC’s 2023–2024 posture — aggressive enforcement against exchanges, issuers, and staking providers — has softened materially through 2025 and into 2026. Public no-action guidance, staff statements, and the withdrawal of several high-profile enforcement matters (including Coinbase, and the pared-back Binance theory) have reset the risk calculus. But softened posture is not repealed law. Howey is still the test. LBRY is still on the books. A T-SAFE-issuing company still has to run the Howey analysis at every step — at the SAFE, at the warrant, at the exercise, at TGE, and again at any secondary-market listing. The instrument does not do the analysis for the company.
How to pick
The short version. If there is no priced equity round on the horizon and the token will exist within twelve to eighteen months, the SAFT still has a narrow use case. If there is a seed or Series A round in view and investors want both equity and token upside, the T-SAFE is the default. If there is a priced equity round happening now and the tokenomics can be paired with the round, the standalone token warrant is the cleanest structure.
The companion worksheet linked above walks through the same decision in seven steps, and provides a comparison table for founders and counsel to work through together. The right answer is almost always the boring one: the instrument that later investors and later diligence teams will recognize on sight and not need to re-argue.
Download: SAFT vs T-SAFE vs Token Warrant — Comparison Worksheet (.docx) — a companion resource for this post. Adapt with counsel before use.
If you are a founder or GC structuring a token-inclusive raise and want a second read on the instrument choice, feel free to reach out to our firm manager, Magda, at Magda@montague.law, or fill out our contact form. Mention you read this post.
