Paid to make a market that was not there: the crypto market-maker cases
Yes for four firms, no for the market. The DOJ says Gotbit served multiple issuers, ZM Quant served Saitama and NexFundAI, and MyTrade ran wash-trade bots for dozens of clients. None of that measures how common the practice is; three firms were charged over offers to wash trade the FBI's own token, NexFundAI.
In video calls between July and August 2024, an employee of a firm registered in the United Arab Emirates explained to a prospective client how its trading bot worked. According to the Department of Justice, the employee said it “basically does self-trades, buying and selling” from multiple wallets so that the result looks like organic trading, and that it was “very hard to track”. The prospective client was a purported crypto company, and the token it wanted promoted, NexFundAI, had been created by the FBI.
That is the fact pattern behind the largest group of cases in this library’s paid crypto market making family. The question this post asks of the record is a simple one. If the same few firms sold the same service to many issuers, then a handful of prosecutions is a window on a practice much wider than the cases. Is that what the charging documents actually say?
What the firms were paid to do
The SEC’s October 2024 complaints describe token promoters who hired “so-called market makers” to provide what the SEC calls market-manipulation-as-a-service: generating artificial trading volume or moving the price of tokens sold to retail buyers. A genuine market maker quotes both sides and takes inventory risk. What was alleged here was trading with no economic purpose, run by a bot, on public exchanges. The mechanic is wash trading; the commercial arrangement is the purchase of it. The technique page sets out where that differs from a venue faking its own volume and from token-loan structures.
The DOJ’s account of what the firms sold is specific. ZM Quant allegedly advertised a “trading bot” that could “create volume”. MyTrade MM offered clients a dashboard on which they set the desired number of daily wash trades, under the label “Volume Support”. Gotbit’s founder allegedly kept spreadsheets comparing “Created Volume” with “Market Volume”. These are allegations in charging documents, apart from what the defendants later admitted, which the next sections set out.
Four firms, and how many clients each is documented to have had
The claim to test is that these were services sold repeatedly. What each source actually supports:
Gotbit. The DOJ says that between 2018 and 2024 Gotbit provided market manipulation services to “multiple cryptocurrency companies” and received tens of millions of dollars in client payments. In its plea Gotbit admitted manipulative trades for clients that included Robo Inu and Saitama. This is the strongest support for the argument, because it is an admission and not only an allegation. The SEC’s matter against Gotbit, filed in October 2024, alleged the scheme in Robo Inu; in August 2026 the SEC filed a proposed settlement with the firm and dismissed its claims against its employee Fedor Kedrov.
ZM Quant. The DOJ says it provided services for multiple companies, including Saitama and NexFundAI. The SEC’s complaint adds SaitaRealty. This is an allegation in charging documents. The library’s own record is the SEC’s October 2024 announcement, and this post has not established a later outcome for ZM Quant.
MyTrade MM. In August 2026 the DOJ said that as of 1 October 2024 the firm was providing its volume-support function to “dozens of clients”, and that the bots it was required to deactivate had been responsible for millions of dollars of daily wash trades in approximately 60 different cryptocurrencies. That is the largest documented client base in the record.
CLS Global. The DOJ said CLS employed over 50 people and that its employee, in the NexFundAI calls, said “we’ve been doing that for many clients”. That is a sales statement by the defendant’s employee, and it is not established by the release as a fact about CLS’s client base.
So the answer for these four firms is that the documents do describe repeat service to several issuers, in Gotbit’s case admitted, in MyTrade’s a matter of the DOJ’s own account of a plea. The same issuer also used more than one firm: the DOJ alleges Saitama’s leadership paid several market makers, including ZM Quant and Gotbit. Saitama, which the DOJ said at its peak had a market value of $7.5 billion, is the common thread.
How the government used its own token
The DOJ’s account is the following. The FBI created NexFundAI, “a purported cryptocurrency company” with a website and an Ethereum-based token that traded on the Uniswap exchange until law enforcement disabled it. Purported promoters of the token held conversations with three market makers: ZM Quant, CLS Global and MyTrade MM. The DOJ says CLS Global agreed to provide market making for the token that included wash trading, and, in a “Market Making proposal”, showed a dashboard of “total volume”, “CLS volume” and “external volume”. CLS traders then bought and sold NexFundAI on Uniswap from CLS’s own trading wallets and reported on the activity created by the volume algorithm. The MyTrade MM founder is described as telling the purported promoters that the firm did self-trades and that its volume bot could be used for pump and dumps.
Because the token was the FBI’s, the promoters in these conversations were government-run. What the firms are alleged to have agreed to was manipulating trading in that token, and in effect the design tested the sales pitch directly: the firms described their methods to a buyer who had asked for volume. What the releases do not say is how many real buyers, if any, bought NexFundAI, or how the FBI ran the token day to day. Those details are not in the documents read for this post, so they are left out.
What became of them
The outcomes as the DOJ reports them: CLS Global pleaded guilty in January 2025 to one count of conspiracy to commit market manipulation and wire fraud and one count of wire fraud, and on 2 April 2025 was ordered to pay $428,059 and put on three years’ probation, prohibited from participating in US crypto markets. Gotbit and its founder Aleksei Andriunin pleaded guilty on 21 March 2025; Gotbit forfeited approximately $23 million in seized cryptocurrency and was placed on five years’ probation during which it must cease to operate, and on 12 June 2025 Andriunin was sentenced to eight months in prison followed by a year of supervised release. On 6 August 2026 MyTrade’s founder, who had pleaded guilty to conspiracy to commit market manipulation and wire fraud, was sentenced to a $10,000 fine.
Individuals other than these founders are in different positions, and the record is thin. As of the DOJ’s 9 October 2024 release, Russell Armand and Max Hernandez had pleaded guilty, Vy Pham had agreed to plead guilty, and Saitama’s chief executive had been arrested in the United Kingdom and was awaiting extradition. The SEC’s parallel release said Armand, Hernandez and Pham had consented to bifurcated settlements, with the court to decide any money. This post has not verified what happened after that date for those individuals, so their current status is not stated here.
The earlier case follows the same shape without a criminal counterpart in this library. In SEC v. Hydrogen Technology, the SEC alleged that Hydrogen hired the South Africa-based firm Moonwalkers in October 2018 to create false market activity with its own trading bot and then sold into it, for profit of more than $2 million. Without admitting or denying the allegations, Hydrogen was ordered to pay disgorgement of $1,516,703.53 with prejudgment interest of $244,531.98 and a $1,035,000 civil penalty, and Hydrogen’s former chief executive the consent judgment for Tyler Ostern, chief executive of Moonwalkers, imposed no penalty because of his cooperation with the SEC.
What the record supports, and what it does not
The argument survives in a narrower form. For four named firms, DOJ and SEC documents describe service to more than one issuer, and one firm admitted it. That is more than a single-case story. What the record does not support is the wider claim that a handful of prosecutions reveals a market-wide practice. Nothing read for this post measures how many other firms offered the service, how many tokens bought it, or how much reported crypto volume was affected. The DOJ’s release headline describes “widespread” fraud and manipulation, and the Acting U.S. Attorney called the investigation the first of its kind; those are the prosecutors’ characterisations, not a measurement.
The library itself is not a sample. It holds what regulators announced, and it has known gaps. The count of cases here is a count of cases here. It is not a prevalence estimate.
The practical reading is narrower and sturdier. A firm’s willingness to describe the method on a sales call, and to explain how to avoid on-chain detection, is what the government tested with NexFundAI. It found three firms willing to discuss it with a buyer the government had invented. That is worth knowing without inflating it into a statistic.
The details above are allegations in charging documents except where a plea, sentence or judgment is stated. Sources: the DOJ release of 9 October 2024 (Eighteen individuals and entities charged), the DOJ releases on CLS Global, Gotbit and MyTrade, and the SEC’s press release 2024-166 and litigation releases 25737, 26154, 26287 and 26598.