Ah, the tween years for chip announcements: Those periods when companies roll out tweaked versions of the big architectural changes we saw over the previous year (or earlier). In AMD’s case, some of the updates to its chip lines were essential. But since it made its big announcements at last CES, what should we realistically expect at CES 2026? The answer is refreshed Ryzen AI lines focused primarily on mobile processors, moving from the 300 series to the 400 series, and a bit more.
Don’t miss any of our unbiased tech content and lab-based reviews. Add as a preferred Google source.
The “bit more” comes in the form of an addition to the Ryzen X3D lineup within AMD’s gaming-focused 9000X series desktop processors. The new Ryzen 7 9850X3D slots in above the Ryzen 7 9900X3D and is essentially the same processor, using slightly better-performing dies from the same pool and boosting the clock speed by 100MHz. In practical terms, the 9850X3D runs at 5.6GHz, compared with 5.5GHz for the 9900X3D. AMD says this translates to roughly a 7% performance improvement.
The Acer Swift Go 16 AI incorporates the Ryzen AI 9 465.
In the same vein, AMD has added two processors to the Ryzen AI Max Plus 300 series: the Ryzen AI Max Plus 392 and Ryzen AI Max Plus 388, which slide in above the plain old Ryzen AI Max 390 and 385, respectively. The one change is to the GPU: It jumps from 32 compute units to 40 CUs, effectively “Plus-ifying” the chips with the 8090S graphics. That upgrade boosts GPU performance for both gaming and AI and provides more affordable alternatives to the Ryzen AI Max Plus 395. People don’t always notice when their CPU is slow, but gamers definitely notice when their GPU is.
AMD Ryzen AI 400 series
| Cores (Zen 5c/Zen 5) | Max boost frequency (GHz) | Total threads | NPU | GPU | GPU compute units | GPU max frequency (GHz) | |
|---|---|---|---|---|---|---|---|
| Ryzen AI 9 HX 475 | 12 (8/4) | 5.2 | 24 | XDNA 2 up to 60 TOPS | Radeon 890M | 16 CU | 3.1 |
| Ryzen AI 9 HX 470 | 12 (8/4) | 5.2 | 24 | XDNA 2 up to 55 TOPS | Radeon 890M | 16 CU | 3.1 |
| Ryzen AI 9 465 | 10 (6/4) | 5 | 20 | XDNA 2 up to 50 TOPS | Radeon 880M | 12 CU | 2.9 |
| Ryzen AI 7 450 | 8 (4/4) | 5.1 | 16 | XDNA 2 up to 50 TOPS | Radeon 860M | 8 CU | 3.1 |
| Ryzen AI 7 445 | 6 (4/2) | 4.6 | 12 | XDNA 2 up to 50 TOPS | Radeon 840M | 4 CU | 2.9 |
| Ryzen AI 5 435 | 6 (4/2) | 4.5 | 12 | XDNA 2 up to 50 TOPS | Radeon 840M | 4 CU | 2.8 |
| Ryzen AI 5 430 | 4 (3/1) | 4.5 | 8 | XDNA 2 up to 50 TOPS | Radeon 840M | 4 CU | 2.8 |
Finally, there’s the Ryzen AI 400 series, which are generally somewhat faster versions of their 300-series equivalents. The HX versions of the 300 series are older — they launched in June 2024 — but they retain mostly the same specs, with only minor speed bumps, such as the 100MHz tweak mentioned earlier. They do get a boost in NPU performance as well, topping out at 55 TOPS and 60 TOPS for the HX 470 and 475, respectively, up from the 50 TOPS found across the rest of the XDNA 2 chips. The remaining models just gain some minimal clock-speed improvements and support for slightly faster memory.
The Ryzen AI Halo
AMD has put together its own compact desktop system specifically for local AI development (up to a 200 billion parameter model) called the Ryzen AI Halo. It’s configured with a Ryzen AI Max chip and 128GB shared memory and supports multiple operating systems; will come preloaded with a host of open source tools and AMD’s ROCm AI API stack.
