Wednesday, November 5th 2025
AMD Readies Ryzen AI MAX+ 388: 8C/16T and Full 40-CU Radeon 8060S GPU
AMD is gearing up to release a new "Strix Halo" Ryzen AI MAX+ SKU specifically designed for gamers who don't need more than 8 cores and 16 threads of CPU power. In a recent PassMark benchmark, a new SKU called Ryzen AI MAX+ 388 was spotted. It features 8 cores, 16 threads, and a fully equipped Radeon 8060S GPU with 40 RDNA 3.5 Compute Units (CUs). Previously, the full Radeon 8060S configuration was only available in the high-end Ryzen AI Max+ 395 model, which combines 16 cores and 32 threads with the 40 CU Radeon 8060S. However, that model is more expensive and better suited for tasks like productivity and AI that benefit from higher core counts.
If and when AMD introduces the 8-core, 16-thread Ryzen AI MAX+ 388, we anticipate that more gaming devices will adopt this chip, as it offers an ideal balance of CPU and GPU power for running modern games. AMD supports up to 128 GB of 256-bit LPDDR5x memory at 8000 MT/s with its Ryzen AI MAX+ SoCs. However, the actual memory configuration will depend on the choices of the OEM or device manufacturer. Some devices, such as SIXUNITED 14-inch tablets, have already been seen using this APU, configured with a 65 W TDP. AMD provides a configurable TDP range of 45-125 W, with a default TDP of 55 W for its top three SKUs. Therefore, we can expect to see various TDP ratings as OEMs tailor this chip to their specific requirements, based on the device's form factor and thermal capacity.
Sources:
@realVictor_M on X, PassMark, via VideoCardz
If and when AMD introduces the 8-core, 16-thread Ryzen AI MAX+ 388, we anticipate that more gaming devices will adopt this chip, as it offers an ideal balance of CPU and GPU power for running modern games. AMD supports up to 128 GB of 256-bit LPDDR5x memory at 8000 MT/s with its Ryzen AI MAX+ SoCs. However, the actual memory configuration will depend on the choices of the OEM or device manufacturer. Some devices, such as SIXUNITED 14-inch tablets, have already been seen using this APU, configured with a 65 W TDP. AMD provides a configurable TDP range of 45-125 W, with a default TDP of 55 W for its top three SKUs. Therefore, we can expect to see various TDP ratings as OEMs tailor this chip to their specific requirements, based on the device's form factor and thermal capacity.


49 Comments on AMD Readies Ryzen AI MAX+ 388: 8C/16T and Full 40-CU Radeon 8060S GPU
This is totally a SKU for gaming handhelds.
The Steamdeck has way lower wattage APU.
We are more likely to see this chip in miniPCs and laptops.
Tryin' to get a 14" lightweight ultrabook with 370 HX - barely 2-3 models to choose from...
Just look at Broadwell and Skylake with GT3e.
This gives me hope on next generations. I mean, I wouldnt buy this but a hypothetical 9060S version with FSR4 support can be godsend for handhelds and smaller laptops.
It would make a killer all-rounder of a laptop but it tends to end up in either ridiculously thin flagship things with insane pricing, coupled with performance strangling cTDPs that leave most of the benefits on the table due to inadequate cooling in such a thin-and-light package. The instant the laptop OEMs make a slightly thicker laptop with better cooling, they immediately switch to Nvidia dGPUs instead. The number of capable AMD APU-only laptops with a decent power budget can be counted on the fingers of one hand for each product cycle. That form-factor best suited to Strix point is also the one that the OEMs seem to ignore the hardest.
They inevitably make their way to mini-PCs, where they're overpriced for the gaming performance their IGPs can provide. Sure, the CPU performance is great, but a 40CU power-limited IGP with performance in the RX9060 ballpark isn't appealing when the cost could have bought you something like a 7600X3D and a 9070XT instead.
videocardz.com/newz/onexfly-apex-ryzen-max-handheld-officially-priced-at-1399-full-spec-model-costs-2299
Maybe with Medusa Halo they'll be more common, since they plan on repurposing that chip for dGPUs too, so there will be a higher production run.
Take it with a grain of salt. Its all hearsay for me too but its what i've come across when searching for a laptop myself.
Besides, if it's like a lot of cases I've seen, both in my hardware and others' experience, dropping 20% of its rated power usually doesn't cause a lot of drop in performance you'd notice unless you were looking at monitoring software.
Still a very interesting chip.
The pricing on 395 max devices rarely make sense for gaming needs. Their saving grace is the 128GB fast (by consumer standards) soldered RAM. There's nothing special about them from an average consumer's perspective.
dGPUs are faster in most scenarios even for laptops.
Minisforum, GMKtek units are available in the market, 16cores and lower memory options should be perfectly fine for gaming.
HP and Asus do make some units but don't quote me on that. Worse pricing though.
If anything, I imagine AMD held off on something that made more sense to gaming because it would cannibalize sales of configurations with a Radeon dGPU. Especially when Notebookcheck is making claims that the 8060S is on par with a laptop 4070 (even though they didn't seem to make it clear said 4070 was on the lower end of the power configuration).