Robert Triggs / Android Authority
Who doesn’t love wild rumor, particularly when it blends high-profile telephones just like the Samsung Galaxy S25 with sizzling subjects like AI? Based on the latest whispers, Samsung’s subsequent flagship telephone shall be powered by a brand new flagship Exynos 2500 processor (along with an as-yet-unannounced Qualcomm Snapdragon 8 Gen 4). The report requires a healthy dose of salt, however the juicy hypothesis is that the chip will function a devoted AI accelerator constructed by Google, not in contrast to the Google TPU discovered within the Pixel 8’s Tensor G3 processor.
For the previous two generations, Samsung has developed its personal AI accelerator for Exynos, consisting of massive and little AI-focused cores (NPUs) and a extra conventional number-crunching digital sign processor (DSP). Introducing a Google-developed processor half could be a tacit admission that Samsung’s AI silicon is behind the competitors, both in efficiency or improvement instruments.
Often, we’d say completely no method, however this isn’t an inconceivable improvement. Google and Samsung have labored carefully on the Tensor chip collection and AI software program like Galaxy AI. Licensing further know-how between the 2 could be enterprise as typical. ASUS already makes use of Google TPUs for its Coral IoT platform, so why not Samsung?
The advantages would, presumably, be stronger AI number-crunching capabilities for the Galaxy S25 collection. Whether or not or not the rumor seems to be true (we’re nonetheless removed from satisfied), it has drawn my consideration to a key inflection level within the present cell AI arms race that’s price diving into.
Somebody will win the AI race, sooner or later
The AI panorama has modified rather a lot up to now yr, and so has the {hardware} necessities to run the newest fashions and instruments. “Primary” early inferencing and cloud-based instruments have shortly given approach to generative fashions that may produce swathes of textual content, pictures, and even video on system, however require vastly extra specialist compute energy to take action. Cellular processor designers have been investing increasingly more transistors into NPU processing elements to maintain up, with Google, Samsung, Qualcomm, MediaTek, and Apple all pursuing barely completely different approaches.
If the rumors are true, Exynos and Tensor could be nearer than ever.
There are all the time winners and losers in an progressive subject, and there’s a component of luck in predicting silicon improvement instructions so a few years prematurely. A part of the issue is that AI workloads range a lot, and determining which directions and bit-depths to help in {hardware} is tough whereas the state of functions stays in flux. The altering panorama could make improvement prices prohibitive, particularly in case you’re not promoting your developments far and vast.
Google’s benefit is having its hand in each AI software program and {hardware} improvement, maybe retaining its TPU improvement slightly extra in tune with altering developments than some others. There are sure to be advantages to utilizing AI {hardware} from the identical firm bringing Gemini Nano fashions to numerous smartphones, in any case.
Robert Triggs / Android Authority
Wanting simply barely additional forward, the way forward for AI functions received’t restricted to a couple proprietary OEM apps that ship along with your telephone (sorry, Galaxy AI). Customers will need to take their AI functions throughout units and platforms like different apps, necessitating third-party developer entry to AI accelerator {hardware}.
Nevertheless, I examined the Exynos and Snapdragon Samsung Galaxy S24 collection AI efficiency numbers utilizing GeekBench ML (admittedly nonetheless in improvement) and located the app couldn’t entry further NPU efficiency by way of the Android NN API. Operating instantly on the CPU and GPU was quicker, which shouldn’t be the case. Maybe this can be a driver subject, or a few of the finest AI telephones in the marketplace don’t seem to present third-party builders quick access to leverage their AI smarts by way of a core Android API.
I ran the identical take a look at on Google’s Pixel 8 Professional and located a notable uplift in some workloads when utilizing Android NN (besides in 32-bit floating level checks, which see no profit). In different phrases, builders can run TensorFlow AI workloads on the Pixel’s TPU for important efficiency beneficial properties.
Operating Ignatov Andrey’s AI Benchmark (outcomes under) showcased a really comparable pattern. Android NN API confirmed mainly no uplift over CPU cores and in addition scored far under Qualcomm’s QNN frameworks on the Galaxy S24 Extremely. Apparently, QNN on the S24 Extremely truly outclassed the Pixel 8’s finest rating by a large margin, highlighting Qualcomm’s AI compute energy. As well as, Qualcomm has its AI Hub instruments for builders to entry its personal on-device AI fashions. Worryingly, Samsung’s ENN possibility couldn’t run on the Exynos Galaxy S24, leaving it caught with a horrible Android NN-based rating.
This isn’t to take subject with the Galaxy S24’s AI efficiency; it clearly works nicely sufficient for Galaxy AI. The purpose is that builders aren’t going to need to implement three or 4 completely different frameworks to realize the perfect AI app efficiency to everybody. It is a broader drawback.
Hopefully, the above has helped spotlight the 2 halves of the AI silicon problem: first creating the {hardware} after which coding the libraries and instruments to open up that {hardware} in a method that builders can and need to construct on. That’s a hefty funding, extra so when silicon is altering so quick.
AI silicon will ultimately coalesce round essentially the most sensible mannequin, and I would not guess in opposition to Google.
If Google has already performed the API legwork work with its TPUs, as seems to be the case, maybe that’s a adequate motive for Samsung to think about Google’s answer over an in-house possibility for future Exynos chipsets, particularly if it believes in a future marketplace for third-party AI apps.
Right this moment’s cell AI panorama is a Wild West, with manufacturers chasing their very own distinctive {hardware} and software program mixtures in a bit to get forward of the competitors. Finally, the market will decide on essentially the most sensible, if not essentially essentially the most highly effective, concepts, and we’ll nearly actually see way more homogeny between units. I wouldn’t guess in opposition to Google right here, even when this shaky AI silicon rumor doesn’t develop into true for the Samsung Galaxy S25 collection.