The TCL QM8L is the latest offering from the TV brand most closely associated with mini-LED backlighting technology. Yes, every other big TV name now makes mini-LED sets, but TCL has been at it the longest, and it manages to make meaningful improvements to its display panels and processing with each new generation of models.

Last year’s TCL QM8K featured the company’s Halo Control System. This, combined with a 23-bit backlight controller and Dynamic Light Algorithm, upped the level of local dimming precision to the point where backlight blooming effects – a key picture quality drawback with LED-based TVs – were nearly non-existent. For the new QM8L, TCL has made further progress in the battle against blooming by adding a 26-bit backlight controller and up to 4,000 local dimming zones (on the 98-inch model), an increase over the QM8K.

The bigger story with the new QM8L series, however, is SQD Mini-LED. SQD stands for Super Quantum Dots, which is an improved version of the Quantum Dot formulation found in many mini-LED TVs, and one that provides extended color gamut coverage (up to 100% BT.2020, according to TCL). The QM8L also features an Advanced Color Purity algorithm and a new Ultra Color filter which uses 5 nanometer particles compared to the 60 nanometer particles in standard mini-LED TVs, and TCL says this combination provides “more accurate pixel-level color” along with more “consistent color saturation.”

The QM8L series isn’t the first TCL TV with Super Quantum Dots – the X11L series the company introduced at CES holds that distinction. The X11L series goes even further than the QM8L on the local dimming front, offering up to 20,000 dimming zones on the 98-inch model, and it also has a super-thin design. With prices starting at $7,000 (for the 75-inch version), the X11L series is about as pricey as TVs get in 2026. How does the more affordable midrange QM8L ($2,999.99 for the 75-inch model I tested) hold up, both against its flagship big brother and last year’s QM8K? Let’s find out.

What Is It?

The TCL QM8L is the step-down series in the company’s SQD-Mini-LED TV lineup for 2026. Aside from Super Quantum Dots and up to 4,000 backlight local dimming zones (2,584 on the 75-inch model I tested), the QM8L series features a high-contrast WHVA 2.0 Ultra Panel with a wide color viewing angle and an anti-reflective screen filter. A new TSR AI Pro processor provides AI-enhanced contrast, color, motion, and upscaling, and there’s Super Resolution 2.0 processing to enhance detail.

HDR support on the QM8L series includes the Dolby Vision, HDR10+, HDR10, and HLG formats, and Dolby Vision 2 Max will be added to that list in a software update planned for summer 2026. Filmmaker Mode and IMAX Enhanced mode picture presets are provided, and the set also has Intelligent picture and sound modes that make automatic adjustments based on content.

Gaming gets a welcome boost on the QM8L series with four HDMI 2.1 ports instead of the two provided on previous TCL TVs. These all support a 4K/144Hz refresh rate, along with 1080p/288Hz for PC gaming. FreeSync Premium Pro is also supported, and TCL’s Game Bar pop-up menu lets you make quick gaming-related onscreen adjustments while playing. TCL also says that the Xbox Game Pass app should soon be available via a software update in May to allow for cloud-based gaming on the QM8L.

TCL’s ZeroBorder design for the QM8L series gives it an almost bezel-free “all-screen” look. Cabinet depth is two inches, but the TV’s tapered side panels give it a slimmer appearance when viewed from the sides. An adjustable height pedestal stand gives you the option to install the screen either flush with a tabletop or elevated to clear space for a soundbar, and there’s a new remote control design with a backlit keypad and built-in mic.

Along with the QM8L’s four HDMI 2.1 ports, there’s an optical digital audio output, USB 2.0 and 3.0 ports, an Ethernet jack and an RF input for an antenna. All connections are located on a side facing panel where they can be easily accessed in a wall-mount installation.

Talking to the TV

The QM8L uses the Google TV smart platform for content browsing and streaming, but this version is enhanced with Gemini interactive AI. That basically means the standard set of Google TV features including apps, Google TV Freeplay, content recommendations, and Live TV are all provided, but there is an additional tab on the home screen that takes you to the Gemini for TV portal. Here you can customize the “voice” of your AI assistant, as well as get a Today’s News brief, create AI-generated images from voice prompts to use as screensavers, and engage in interactive discussions with Gemini on a wide range of topics.

According to TCL, a future update to Gemini interactive AI on the QM8L will additionally let you adjust picture and sound settings via voice commands, as well as generate custom videos using the Veo on Google TV feature.

I rarely use voice commands to control my TV or any other device, and I generally don’t bother with AI. To test this feature on the QM8L, however, I asked “Where are old growth forests located on the east coast of the US?” and received a surprisingly detailed list in an AI summary. Gemini is conversational, so I asked further questions about specific locations on the list and even nearby places to stay overnight. This being a Google product, there were of course YouTube videos provided for each listing, and these let me get a better look at the forest sites.

The QM8L features an ATSC 3.0 tuner, and broadcast channels pulled in by a connected antenna can be browsed separately in grid-guide format by selecting the Antenna TV tab in the Live TV portal. This feature lets you easily access your local stations and browse their programming options separately from the national network broadcasts carried by Freeplay.

Audio Unlimited

An Audio by Bang & Olufsen speaker array on the QM8L delivers Dolby Atmos soundtracks via down-firing speakers located on the TV’s bottom surface and there are also two “subwoofers” located on the back. The built-in sound quality is good overall, with clear dialogue, clean-sounding bass, and decent enough height effects, especially when the Vertical Sound Field setting in the Acoustics Laboratory menu is set to Medium or higher. Other options here include Auto Volume Control and AI Sonic-Adaption, a feature that emits test tones to analyze your viewing environment and applies room correction EQ. There is also a Bluetooth personal audio option for viewing with wireless headphones.

You’ll of course get better sound by using a soundbar or other external speaker system and TCL provides plenty of options here. At minimum you can buy a subwoofer from TCL and pair it wirelessly with the TV for enhanced bass. The QM8L also supports Dolby Atmos FlexConnect, with the option to pair up to four satellite speakers and a subwoofer for a 4.1.4-channel system. Your options here are limited to TCL’s Z100 wireless speaker and Z100-SW subwoofer, but the simple setup and installation flexibility that FlexConnect provides makes it worth consideration.

Setup & Viewing Impressions

I started out my testing of the QM8L by making measurements using Portrait Display’s Calman Color Calibration software in the set’s default Filmmaker Mode and Standard presets with Adaptive Brightness disabled. Peak HDR brightness measured on a white 10% window pattern in Filmmaker Mode was 1,992 nits and 708 nits on a 100% (fullscreen) white pattern. In Standard, peak HDR brightness was 1,938 and 719 nits, respectively, on 10% and fullscreen patterns.

I should note here that TCL cites up to 6,000 nits peak brightness in its specs for the QM8L series. And while I wasn’t able to duplicate that with any variation of settings, I did measure up to 4,500 nits on initial measurements, though the actual brightness proved to be significantly lower after a few seconds once the display stabilized on the test pattern. The brightness measurements listed above reflect the QM8L’s stabilized brightness levels.

The QM8L’s color gamut coverage in Filmmaker Mode measured 89.3 for BT.2020 and 97.7 for P3. Those results are an improvement on last year’s QM8K, especially for BT.2020, and they only slightly trail what I measured on the Samsung R95H Micro RGB TV. 

Delta-E (the margin of error between the test pattern source and what’s displayed on-screen) averaged 4.9 for grayscale and 3.3 for color, both of which are higher than the 3.0 result considered to be the threshold for what’s indistinguishable from perfect to the human eye. To calibrate the TV, I changed the Color Temperature slider in the color menu from its default 5 setting to 4, and then made further adjustments in the TV’s 2-point White Balance menu. Post-calibration, the measured dE averaged 0.7 for grayscale and 1.1 for color.

For gaming, selecting the QM8L’s Game picture preset and turning on the Game Master mode in the System menu settings will result in the lowest input lag. Using a Leo Bodnar input lag meter, I measured 9.8ms for a 4K 60Hz input – a  much improved result over previous TCL TVs I’ve tested, all of which averaged in the 13-14ms range.

A look at the demonstration reel and Starfield animations from the Spears & Munsil Ultra HD Benchmark test disc revealed virtually no backlight blooming on the QM8L. Other tests showed that TCL’s TV had very good off-axis uniformity, with colors maintaining saturation even when viewed at a far off-center seat. The set’s contrast-enhancing anti-reflective screen coating also proved effective at reducing screen glare when viewing in a room with bright overhead lighting, though it didn’t fully eliminate reflections.

Watching movies and TV, what I found most noteworthy about the QM8L’s picture was that the colors were rich and fully saturated, but not to the point of looking cartoonlike. Playing a 4K Blu-ray of Spider-Man: Into the Spider-verse, the animated movie’s colors in the sequence where Spider-Man battles the Green Goblin looked about as vivid as I can recall ever seeing on a TV. The QM8L’s strong contrast and refined shadow detail also gave the picture a strong sense of depth in this Dolby Vision HDR transfer.

Alien: Earth on Hulu (watched in 4K/HDR10) has impressive visual effects for a TV series, and they give the scenes in the chambers and laboratories of the interplanetary vessels a high level of realism. Same as with Into the Spider-verse, shadows in dark scenes showed a high degree of detail, and bright highlights were also powerful, yet detailed. But it wasn’t just sci-fi dramas that looked great on the QM8L. When I watched medical drama The Pitt in 4K/Dolby Vision on HBO Max, the emergency room scenes had an alluring level of brightness, and the skintones of the multiethnic cast all looked natural.

The QM8L’s motion handling was for the most part good, although I did note a fair amount of judder artifacts when watching a scene in the movie No Time to Die where James Bond traverses a hillside cemetery. Setting both the Judder and Blur settings to 3 in the Custom Motion fixed this, however, and I didn’t note any “soap opera” effect after making those adjustments.

The Bottom Line

TCL’s QM8L offers a marked improvement over last year’s also very impressive QM8K series. It gets a visible, and measurable upgrade in color performance through the introduction of Super Quantum Dots, and its Halo Control System with a 26-bit backlight controller and up to 4,000 local dimming zones mean that backlight blooming artifacts are nearly non-existent. The QM8L series is also a great TV for gaming, with 4K/144Hz support across four HDMI 2.1 ports, ultra-low input lag in Game mode, and soon-to-come Xbox Game Pass cloud gaming support. 

The Google TV with Gemini feature is a great addition for this series, and it makes content searches, and also general research on most any topic, a breeze. TCL carries over the QM8K’s sleek design to this series as well, and the TV’s impressive built-in sound is augmented by Dolby Atmos FlexConnect 4.1.4 support. 

At $3,000 for the 75-inch model I tested – the same price as last year’s QM8K series TV and $4,000 less than the brighter but otherwise similarly featured X11L flagship – the TCL QM8L is a great value for all that it offers. The high-end TV market in 2026 is going to be dominated by RGB LED and flagship OLED models, and you can bet they will all be pricey. Compared to those, and also to TCL’s own SQD-Mini-LED flagship, the QM8L is a relative bargain.

Pros:

• High brightness
• SQD tech brings extended P3 and BT.2020 color gamut coverage
• Refined local dimming
• Impressive off-axis color uniformity
• Anti-relection screen
• Google TV with Gemini AI assistant
• Audio by Bang & Olufsen speaker array
• Dolby Atmos FlexConnect support
• Wireless subwoofer support
• Four HDMI ports with 144Hz support
• Ultra-low input lag for gaming
• ATSC 3.0 tuner

Cons:

• Slightly high color point and color temperature errors in Filmmaker Mode

Where to buy:

Related Reading:

Summary: Anthropic CEO Dario Amodei is meeting White House Chief of Staff Susie Wiles on Friday to negotiate access to Mythos, a frontier AI model that can identify and exploit thousands of zero-day vulnerabilities across every major operating system and browser. The meeting follows Anthropic’s blacklisting by the Pentagon after Amodei refused to remove safety restrictions, and comes as US Treasury, the intelligence community, CISA, and UK financial regulators all seek access to the model through Anthropic’s controlled Project Glasswing programme.

Anthropic CEO Dario Amodei is scheduled to meet White House Chief of Staff Susie Wiles on Friday in what represents the most significant step yet toward resolving the company’s standoff with the Pentagon over its refusal to remove safety restrictions from its AI models. The meeting comes as multiple US government agencies, including the Treasury Department, the intelligence community, and the Cybersecurity and Infrastructure Security Agency, are seeking access to Anthropic’s Mythos model, a frontier AI system whose cybersecurity capabilities have triggered emergency briefings from Washington to London to Ottawa.

Mythos, announced on 7 April, is not a cybersecurity product. It is a general-purpose AI model that, during testing, turned out to be capable of identifying and exploiting thousands of previously unknown zero-day vulnerabilities across every major operating system and web browser. It found flaws that had survived decades of human security review and millions of automated tests. When directed to develop working exploits, it succeeded on the first attempt in more than 83% of cases. It is the first AI model to complete a 32-step corporate network attack simulation from start to finish.

Anthropic chose not to release Mythos publicly. Instead, it created Project Glasswing, a controlled access programme that provides the model to roughly 40 vetted organisations, including Amazon Web Services, Apple, Google, Microsoft, Nvidia, Cisco, CrowdStrike, JPMorgan Chase, and Palo Alto Networks, to find and fix vulnerabilities in critical software before they can be exploited. The company has committed up to $100 million in Mythos usage credits and $4 million in donations to open-source security organisations.

The Pentagon conflict

The White House meeting is the product of a dispute that has escalated since February. Defense Secretary Pete Hegseth demanded that Anthropic grant the Pentagon unfettered access to its models across all lawful purposes, including potential use in autonomous weapons systems and domestic surveillance. Amodei refused. Hegseth designated Anthropic a national security supply-chain risk, a label previously reserved for companies associated with foreign adversaries, effectively blacklisting it from government contracts.

Anthropic sued the Trump administration in early March, filing two federal lawsuits alleging illegal retaliation. A federal judge initially blocked the blacklisting, but an appeals court reversed that decision on 8 April, leaving Anthropic excluded from Department of Defense contracts while litigation continues. The company can still work with other government agencies.

The paradox is that the same government that blacklisted Anthropic now wants access to its most powerful model. The Treasury Department is seeking Mythos to hunt for vulnerabilities in its own systems. Parts of the intelligence community and CISA are already testing it. The White House Office of Management and Budget is setting up protections to allow federal agencies to use a controlled version. Axios reported that Anthropic has hired Trumpworld consultants to facilitate negotiations, and that Friday’s meeting is designed to pave the way toward a deal.

Why Mythos matters

JPMorgan Chase CEO Jamie Dimon said publicly that Mythos “reveals a lot more vulnerabilities” for cyberattacks. The UK’s AI Security Institute evaluated a preview version and found it “substantially more capable at cyber offence than any model previously assessed,” noting that it is the first model capable of chaining multiple attack steps into complete intrusions end to end. The Council on Foreign Relations called it “an inflection point for AI and global security.”

The defensive case for Mythos is straightforward: if an AI model can find vulnerabilities that human security teams and automated testing have missed for decades, giving that model to the organisations responsible for defending critical infrastructure lets them fix the holes before adversaries discover them. The offensive risk is equally straightforward: the same capability in hostile hands would be catastrophic. Anthropic’s decision to restrict access rather than release publicly is a direct application of the safety principles that put it in conflict with the Pentagon.

The company’s commercial trajectory gives it leverage in the negotiation. Anthropic’s annualised revenue has reached $30 billion, it has attracted investor offers at an $800 billion valuation, and it is exploring an IPO. It does not need Pentagon contracts to survive. What it needs is a resolution that preserves its safety commitments while restoring its ability to work with the broader US government, a position that the Wiles meeting is designed to explore.

The global response

Mythos has become a subject of concern well beyond Washington. The Bank of England’s Governor Andrew Bailey named it explicitly as a cybersecurity risk in a speech at Columbia University on 15 April. The Bank’s Cross Market Operational Resilience Group is convening an emergency briefing within the fortnight with the CEOs of the UK’s eight largest banks, four financial infrastructure providers, two insurers, and representatives from the Treasury, the Financial Conduct Authority, and the National Cyber Security Centre.

Anthropic is planning to provide Mythos access to select British banks within days as part of Project Glasswing’s expansion, and is quadrupling its London office to 800 staff in King’s Cross. The UK’s AI Security Institute, which has an existing evaluation partnership with Anthropic, published its technical assessment on 17 April. Canadian Finance Minister François-Philippe Champagne described Mythos as an “unknown unknown” being discussed at IMF meetings. Global regulators are coordinating on how to assess and manage the cybersecurity implications.

The geopolitical dimension is unavoidable. The US government’s desire for Mythos access exists in tension with its punishment of the company that built it. Anthropic’s willingness to provide the model to UK banks and regulators while locked in litigation with the Pentagon creates a situation in which America’s closest ally may have access to a critical national security tool before its own government does. That dynamic gives the White House an incentive to resolve the dispute that transcends the original disagreement over safety guardrails.

What a deal might look like

The outlines of a potential resolution are visible. Anthropic would restore its eligibility for government contracts and provide Mythos access for defensive cybersecurity purposes. The Pentagon would withdraw the supply-chain risk designation. Anthropic would maintain its restrictions on autonomous weapons and mass surveillance applications but potentially agree to a process for reviewing specific military use cases that do not cross those lines. Both sides have reasons to compromise: Anthropic because the blacklisting damages its enterprise credibility, and the administration because it needs the technology.

Whether Amodei and Wiles reach that kind of arrangement on Friday or simply begin the process of getting there is less important than what the meeting represents. The company that built the most capable cybersecurity tool in existence did so as a byproduct of building a general-purpose AI model, then restricted its release on safety grounds, then was punished by the government for maintaining those same safety principles, and is now being courted by that government because the tool is too valuable to ignore.

That sequence captures something essential about where AI governance stands in April 2026. The technology is advancing faster than the institutions responsible for managing it can adapt, and the companies that take safety seriously are simultaneously rewarded by the market and penalised by the state. Mythos is the sharpest example yet of a model whose capabilities are so consequential that restricting it and releasing it are both defensible positions, and the argument between them is playing out not in a research paper or a congressional hearing but in the West Wing.

• AI models now perform strongly in obscure languages with minimal training data
• Cross-lingual transfer allows shared patterns to boost rare language performance
• Tokenizer efficiency improvements significantly impact multilingual processing cost and quality

Large language models (LLMs) are closing the global language gap at an unexpected pace, with frontier models now performing well in rare languages that previous generations struggled with.

According to RWS’s TrainAI Multilingual LLM Synthetic Data Generation Study, Google‘s Gemini Pro achieved high-quality scores above 4.5 out of 5 in Kinyarwanda, a language spoken by about 12 million people in Rwanda, Uganda, and the DRC.

More Information

Log-Periodic Dipole Array (LPDA) fed parabolic reflector antennas serve high-gain broadband applications such as satellite communication, radio astronomy, and wideband radar, where constant performance over wide frequency ranges is essential. Despite nearly seven decades of research since the pioneering work of DuHamel and Ore in 1958, the synthesis and analysis of these antennas remain complex, requiring the tuning of many parameters across broad bandwidths. Traditional simulation approaches — combining Method of Moments for the LPDA with physical optics for the reflector — cannot account for mutual coupling between the feed and dish, and break down when support struts or electrically large reflectors are involved. This white paper presents an advanced full-wave simulation methodology using higher order basis functions that reduces unknowns by an order of magnitude compared to conventional MoM, demonstrates a practical three-step design strategy from stand-alone feed optimization through full antenna integration, and provides validated results for reflector diameters ranging from 24.2 λ to 242 λ with bandwidth ratios of 10:1 — all executed on standard desktop hardware with CPU/GPU acceleration.

from the his-own-people-are-still-the-worst dept

Just to be clear, when I refer to “Trump” in terms of his administration, I’m referring to the collective hive mind of dangerous enablers he employs. Trump, by himself, is incapable of closing an umbrella. It’s the people around him that are dangerous, since they’re able to convert his rants and brain stem impulses into action.

While it’s understandable that an aspiring autocrat like Trump would feel threatened by a movement dedicated to opposing fascists, it’s only now that he’s returned to office that he can do anything about it. Deliberately ignoring the fact that the most dangerous domestic terrorists are located on the far right of the political spectrum (including the hundreds of people he pardoned for assaulting police officers and raiding a federal building following his 2020 election loss), Trump’s administration is once again attempting to turn protected First Amendment activity into terroristic acts worthy of lengthy minimum federal sentences.

The United States was as concerned as always about Islamist terrorism, said the official, Monica A. Jacobsen, according to a copy of her prepared remarks reviewed by The New York Times and three officials briefed on the meeting. But, she told her counterparts from Europe, Canada and Australia, the Trump administration also wanted more attention on what it believed was an insidious, underestimated threat: the far left.

Western governments must combat “antifa and far-left terrorism,” Ms. Jacobsen’s prepared remarks asserted, casting the effort as an evolution in counterterrorism following the “global war on terror.” Her prepared speech defined far-left terrorism to include threats from communists, Marxists, anarchists, anticapitalists and those with “eco-extremist” and “other self-identified antifascist ideologies.”

“As always” is a nice touch. It’s always a good idea to keep an “Islamist” scapegoat in the yard, especially when you’re busy losing a war with Iran. Not only does it generate steady work for bored FBI agents, but it also allows Trump to continue pretending the mass deportation of hardworking, tax-paying non-whites is somehow contributing to the effort to root out an alleged “1,700 Iran sleeper cells” in the United States. (No “sleeper cell” has been broken up or deported despite Trump claiming the government already knows who these “sleepers” are and where they’re located.)

As evidence of the dangerousness of “far left” terrorists, Jacobsen pointed to a single protest in Milan, Italy, in which police and protesters “clashed” — the favorite euphemism deployed by people who wish readers to believe protesters were just as violent as law enforcement officers.

Meanwhile, the administration can’t actually find any hard evidence to back up its assertions about the supposed violent threat posed by far left activists.

In November, the State Department took the first major step in the strategy by designating four leftist groups in Europe — two in Greece, one in Germany and another in Italy — as terrorist organizations. None of the groups has been known to have plotted attacks on Americans in the past decade, which is usually a criterion for such a designation.

All it can do — as it did late last year when Trump unilaterally declared “antifa” a terrorist organization — is disappear any evidence to the contrary — something it did less than 48 hours after Charlie Kirk was killed by gunman during a campus appearance.

Even if you were to decide that what’s being claimed about “far left terrorism” in Europe by this administration is somehow true, you can’t ignore the facts on the ground here in the United States:

Over the past decade, right-wing extremists have killed 112 people across 152 terrorist attacks in the United States, according to an analysis by the Center for Strategic and International Studies, a bipartisan research institution. Over the same period, left-wing extremists killed 13 people over 35 attacks, according to the analysis, while jihadist attacks left 82 dead.

Even if Trump hadn’t spent his entire term so far routinely insulting and berating our European allies, it’s still unlikely he would have been able to convince them to ignore the reality of the situation for the sole purpose of future abuses of civil liberties and rights.

Trump has been pounding this table since late last year, but now he’s finding fewer world leaders willing to indulge his fantasies or nod politely as Trump’s emissaries literally make shit up about left-wing activist groups.

The State Department wants to bring foreign law enforcement officials from at least 17 countries to The Hague in May for a workshop on how to fight far-left groups like antifa.

[…]

Formal invitations had not been sent as of last week, in part because Congress had to approve funding. U.S. officials told The Times that foreign governments had expressed less interest in the events than the State Department had hoped.

Advertisement

Once again, let’s pause to reflect on these claims. “Antifa” simply stands for “anti-fascist.” You barely have to move left at all to oppose fascism. All you would have to do is move to the left just far enough to align with… I don’t know… Ronald Reagan? And yet this administration is so stupid and thuggish that it actually thinks it can portray people opposed to fascism as more dangerous than US citizens who actively support it.

Everything else on Trump’s list of “domestic threats” is just a lazy rip-off of McCarthyism. “Far left” supposedly covers Communists, Marxists, “anti-capitalists” (yet another tell), and “eco-extremists.” In other words, people who disagree with this particular president and his policies. Free speech is what it is. But Trump and his enablers want people to go away for decades by turning dissent into terrorism.

Meanwhile, the true terrorist threat that is the extremely foreseeable result of the war in Iran is being back-burnered in favor of locking up people who just want to see this country remain a democratic republic. Fortunately for us, the rest of the world is no longer interested (Israel, Hungary, and Russia aside) in pretending Trump poses less of a threat than the people he wants to punish.

Filed Under: antifa, domestic terrorists, donald trump, eu, far right extremists, fascism, free speech, insurrection, january 6, trump administration, white christian nationalists

Kyrgyzstan-based cryptocurrency exchange Grinex has suspended its operations after suffering a $13.7 million hack attributed to Western intelligence agencies.

The funds were stolen from cryptocurrency wallets belonging to Russian users, as the platform enables crypto-ruble exchange operations between Russian businesses and individuals.

Launched early last year, Grinex has Russian links and is believed to be a rebrand of Garantex, a Russian crypto exchange whose admin was arrested and whose domains were seized over allegations of processing more than $100 million in illicit transactions and enabling money laundering.

In August 2025, the U.S. Department of the Treasury announced sanctions against Grinex, based on evidence that the exchange service was a continuation of Garantex activity, accepting the same actors, their funds, and facilitating an identical role as an illegal operations enabler.

Grinex continued to operate, providing Russia with some level of financial sovereignty and ability to bypass international sanctions that impacted banking and transactions, mainly through a Russian ruble-backed stablecoin named A7A5, which was directly adopted from Garantex.

The exchange says that the type of attack and the digital footprint indicate a threat actor associated with “foreign intelligence agencies” that have “an unprecedented level of resources and technology, accessible only to entities of hostile states.”

“According to preliminary data, the attack was coordinated with the aim of directly harming Russia’s financial sovereignty,” Grinex states.

Blockchain analysis firm Elliptic reports that the theft occurred on Wednesday at 12:00 UTC, and the stolen funds were sent to TRON and Ethereum addresses, then converted into TRX and ETH through the SunSwap decentralized trading protocol.

TRM Labs identified 70 attacker addresses and also discovered a second hack at TokenSpot, another exchange based in Kyrgyzstan with ties to Grinex.

TRM Labs links TokenSpot to Houthi-linked laundering operations, weapons procurement, and the InfoLider influence operation in Moldova, all aligning with Russian strategic goals.

Neither Grinex’s announcement nor Elliptic’s or TRM Labs’ reports provides any evidence pointing to a specific perpetrator, and no technical evidence or indicators were provided to support the exchange’s attribution to Western intelligence services.

BleepingComputer has contacted Grinex about attribution of the attack, but we have not received a response by publishing time.

AI chained four zero-days into one exploit that bypassed both renderer and OS sandboxes. A wave of new exploits is coming.

At the Autonomous Validation Summit (May 12 & 14), see how autonomous, context-rich validation finds what’s exploitable, proves controls hold, and closes the remediation loop.

Advertisement

Claim Your Spot