The Xiaomi 15 Ultra has just surfaced, and it has an odd-looking camera layout. These images comes from SmartPrix, a site that partnered with Yogesh Brar, a tipster.

Do note that the tipster in question here has a mixed track record, but this is not the first time this design has appeared. It surfaced earlier this month in a sketch form, and this is only its render form.

The Xiaomi 15 Ultra has just appeared with an odd-looking camera layout, but these are not official renders

It’s obvious that the renders included here have been created by a third-party, they’re not official leaks. So it’s possible they’re simply based on the sketch that surfaced earlier this month.

In any case, you will notice that the phone has rounded corners and a centered display camera hole. The bezels are very thin, and even though they don’t look uniform in this render, chances are that they will be.

All the physical buttons sit on the right-hand side of the phone. We don’t see its frame on the side, but it’ll likely be mostly flat. On the back, you’ll spot a huge camera oreo, which takes up a huge chunk of the phone’s upper side.

Inside that camera oreo, four cameras are located. The main one sits at the bottom, and it’s centered between two other shooters. The periscope telephoto camera is included in the top-right corner.

Leica will once again be a part of the package

The Leica logo is also visible on the inside of this camera oreo, and the same goes for an LED flash. The camera island shown here will likely protrude quite a bit on the back of the device. Xiaomi will use larger camera sensors.

The source pictured this phone in black and white colors here, with a vegan leather backplate., It’s quite possible that the phone will arrive in those colors, with that material on the back. Just note that these are not leaked renders, so these are not color leaks either. The real thing will likely look at least a bit different.

Unlike the Xiaomi 15 and Xiaomi 15 Pro, the Xiaomi 15 Ultra is not expected to arrive this month. Those two phones are coming on October 29, while the Xiaomi 15 Ultra will follow next year.

The popcorn bucket wars just became literal with Cinemark’s latest entry for Gladiator II. The theater chain’s new entry is not only shaped like the Roman colosseum, it plays a cutesy augmented reality gladiator battle when you point your smartphone at QR code on the bottom (AR-ENA, get it?). The butter on the popcorn is Cinemark’s tagline, claiming “you can eat war.”

In fact, all of the ad copy is brilliantly cheesy: “Every kernel of strength, every ounce of honor, is for the glory of Rome. As you preside over this gladiator arena, you can… eat war. Finish the popcorn and unleash the battle within. You will be entertained.” Being intoned in the Honest Trailer style takes it up an extra notch.

It’s the latest popcorn bucket movie merch, following high-profile entries from Dune and Deadpool. We’ve also seen entries for Ghostbusters: Frozen Empire, Beetlejuice Beetlejuice and others that didn’t quite capture the same zeitgeist.

Sure, you might need to wipe off some popcorn grease to read the Cinemark bucket’s QR code, and the AR animation of two fighting gladiators is reminiscent of a PlayStation 2 render. Still, neither the Dune nor Wolverine buckets boast any interactive features, so the Gladiator II bucket has them beat there — and it’s a smart way to rope in the tech press.

Cinemark’s vessel is also plausibly shaped like a popcorn bucket with its colosseum form. The same can’t be said for Dune‘s sandworm-shaped bucket or Deadpool’s Wolverine head bucket (Dune director Denis Villeneuve called the latter “horrific” and he’s right). If you’re looking to expand your collection, the Gladiator II popcorn bucket will arrive “soon” and the movie itself hits theaters on November 22.

A pair of researchers at OpenAI has published a paper describing a new type of model — specifically, a new type of continuous-time consistency model (sCM) — that increases the speed at which multimedia including images, video, and audio can be generated by AI by 50 times compared to traditional diffusion models, generating images in nearly a 10th of a second compared to more than 5 seconds for regular diffusion.

With the introduction of sCM, OpenAI has managed to achieve comparable sample quality with only two sampling steps, offering a solution that accelerates the generative process without compromising on quality.

Described in the pre-peer reviewed paper published on arXiv.org and blog post released today, authored by Cheng Lu and Yang Song, the innovation enables these models to generate high-quality samples in just two steps—significantly faster than previous diffusion-based models that require hundreds of steps.

Song was also a leading author on a 2023 paper from OpenAI researchers including former chief scientist Ilya Sutskever that coined the idea of “consistency models,” as having “points on the same trajectory map to the same initial point.”

While diffusion models have delivered outstanding results in producing realistic images, 3D models, audio, and video, their inefficiency in sampling—often requiring dozens to hundreds of sequential steps—has made them less suitable for real-time applications.

Theoretically, the technology could provide the basis for a near-realtime AI image generation model from OpenAI. As fellow VentureBeat reporter Sean Michael Kerner mused in our internal Slack channels, “can DALL-E 4 be far behind?”

Faster sampling while retaining high quality

In traditional diffusion models, a large number of denoising steps are needed to create a sample, which contributes to their slow speed.

In contrast, sCM converts noise into high-quality samples directly within one or two steps, cutting down on the computational cost and time.

OpenAI’s largest sCM model, which boasts 1.5 billion parameters, can generate a sample in just 0.11 seconds on a single A100 GPU.

This results in a 50x speed-up in wall-clock time compared to diffusion models, making real-time generative AI applications much more feasible.

Reaching diffusion-model quality with far less computational resources

The team behind sCM trained a continuous-time consistency model on ImageNet 512×512, scaling up to 1.5 billion parameters.

Even at this scale, the model maintains a sample quality that rivals the best diffusion models, achieving a Fréchet Inception Distance (FID) score of 1.88 on ImageNet 512×512.

This brings the sample quality within 10% of diffusion models, which require significantly more computational effort to achieve similar results.

Benchmarks reveal strong performance

OpenAI’s new approach has undergone extensive benchmarking against other state-of-the-art generative models.

By measuring both the sample quality using FID scores and the effective sampling compute, the research demonstrates that sCM provides top-tier results with significantly less computational overhead.

While previous fast-sampling methods have struggled with reduced sample quality or complex training setups, sCM manages to overcome these challenges, offering both speed and high fidelity.

The success of sCM is also attributed to its ability to scale proportionally with the teacher diffusion model from which it distills knowledge.

As both the sCM and the teacher diffusion model grow in size, the gap in sample quality narrows further, and increasing the number of sampling steps in sCM reduces the quality difference even more.

Applications and future uses

The fast sampling and scalability of sCM models open new possibilities for real-time generative AI across multiple domains.

From image generation to audio and video synthesis, sCM provides a practical solution for applications that demand rapid, high-quality output.

Additionally, OpenAI’s research hints at the potential for further system optimization that could accelerate performance even more, tailoring these models to the specific needs of various industries.

Lithuania’s Vinted has secured a new valuation of €5 billion (around $5.4 billion at current exchange rates), after the second-hand fashion marketplace closed a secondary share sale worth €340 million ($367 million).

The transaction was led by private equity giant TPG, with other new participants including Baillie Gifford, FJ Labs, Hedosophia, Invus Opportunities, Manhattan Venture Partners, and Moore Strategic Ventures. It’s unclear how much Vinted’s existing investors cashed out, but the company says that all its existing institutional investors — which include Accel, EQT, Insight Partners, and Lightspeed Venture Partners — have retained at least some stake.

It’s proving to be a bumper year for secondary market transactions, particularly in Europe, as scale-ups seek to unlock liquidity for their employees and VCs in a decidedly tepid IPO market. In the past few months alone, we’ve seen neobanks Revolut and Monzo pursue secondary market routes, attaining lofty valuations off the back of strong user growth and profitability.

In the U.S., meanwhile, fintech giant Stripe followed a similar path to unlock liquidity, reaching a private valuation of $65 billion back in February as it continues to delay a long-rumored IPO. This figure later jumped to $70 billion as Sequoia sought a larger stake from existing investors.

Vinted CEO Thomas Plantenga (pictured above) noted that the sale “rewards our employees for their dedication in making Vinted a success.” The company was valued at €3.5 billion ($3.8 billion) pre-money for its previous €250 million Series F fundraise back in 2021. Since then, it has gone from strength to strength, reporting record revenue growth of 61% in 2023 compared to the previous year and reaching profitability for the first time.

At the same time, Vinted has expanded geographically and is also extending beyond its core fashion roots into the electronics realm — a growth trajectory that prompted marketplace stalwart eBay to respond by removing seller fees in key European markets.

Three industry groups are suing to prevent the Federal Trade Commission (FTC) from enforcing its new “Click to Cancel” rule that requires companies to make it easy to cancel subscriptions, according to Reuters. And yes, it’s exactly who you’d expect.

Click to cancel expands the Negative Option Rule to forbid businesses from making customers cancel services using a method that differs from how they signed up. So, if you sign up online, you must be allowed to cancel online, rather than needing to call a support line, write a letter, or show up in person. Most aspects of the rule, assuming it isn’t blocked, will go into effect 180 days from its entry into the Federal Register.

That’s “arbitrary, capricious, and an abuse of discretion,” the Internet and Television Association, Electronic Security Association, and Interactive Advertising Bureau allege in their complaint filed with the US Fifth Circuit Appeals Court today. The groups — many of whose member companies profit from subscriptions that are easy to start and harder to stop — argue that the FTC is trying to “regulate consumer contracts for all companies in all industries and across all sectors of the economy.”

Indeed, the rule applies to any automatically renewing subscription, whether it’s a gym membership or Amazon Prime, including free trials or those plans that ship you easy-to-cook dinners. The horror!

Arrowhead Game Studios has confirmed that Helldivers 2 will receive a PS5 Pro upgrade in the future.

That’s according to Arrowhead community manager ‘Twinbeard’ who revealed on the Helldivers 2 Discord channel that the popular third-person online shooter will eventually get some form of PS5 Pro upgrade at a later date, but stopped short of sharing what those enhancements will be (via PSU).