What the “Ryzen AI 9 HX 370” Name Tells Us
When AMD first introduced the Ryzen AI branding at CES 2023, the company signaled a new direction for its mobile line‑up: integrating a dedicated neural‑processing unit (NPU) alongside the traditional Zen CPU cores. The “HX” suffix has been used for AMD’s high‑performance laptop chips (for example, the Ryzen 9 6900HX and Ryzen 9 7945HX), while “370” appears to follow the naming pattern used for the 300‑series of the Radeon RX 6000 GPUs, hinting at a possible partnership with a newer graphics architecture.
In short, the name suggests a flagship mobile processor that pairs AMD’s most powerful Zen 4‑based cores with an AI‑focused accelerator, and likely a next‑gen Radeon RX 7600M XT or similar GPU. While AMD has not officially announced a product called “Ryzen AI 9 HX 370,” the combination of known naming conventions and recent roadmap hints gives us a solid basis for discussing what such a chip could look like.
The CPU Core Architecture: Zen 4+ with Higher Clock Limits
All recent “HX” chips have been built on the Zen 4 core architecture, delivering up to 8 cores and 16 threads in the Ryzen 9 7945HX. The “AI” variant adds an NPU but otherwise keeps the same core count. Assuming the “370” is an evolution rather than a brand‑new design, we can expect:
- Eight Zen 4 cores, each capable of 2‑way simultaneous multithreading (SMT) for 16 threads total.
- Base clock speeds in the low 2.0 GHz range, with boost frequencies potentially reaching 5.2–5.4 GHz under light loads, similar to the Ryzen 9 7945HX.
- Improved power management, leveraging AMD’s “Precision Boost Overdrive” (PBO) 2.0 to dynamically adjust voltage and clock rates for sustained performance.
These figures align with what AMD has shown in its 2024 roadmap for “next‑gen mobile” CPUs, meaning the “370” would sit at the top of the performance envelope for laptop processors.
The AI Engine: A Dedicated NPU on‑chip
AMD’s AI push hinges on a purpose‑built NPU that sits alongside the CPU cores and the integrated Radeon graphics block. In the Ryzen 7 7840U, the NPU offers 2 TOPS (tera‑operations per second) of AI compute, enough for real‑time video up‑scaling, voice assistants, and on‑device inference for machine‑learning models.
For a “Ryzen AI 9 HX 370,” the expectations are higher:
- At least 4 TOPS of AI throughput, roughly double the 7840U’s capability.
- Support for common AI frameworks such as TensorFlow Lite, ONNX Runtime, and AMD’s own Radeon AI Engine SDK.
- Hardware acceleration for video‑enhancement pipelines (e.g., AMD’s “AI‑SuperResolution”), speech recognition, and computer‑vision tasks.
AMD has emphasized that the NPU operates independently of the CPU cores, meaning that AI workloads can run without draining the main processor’s thermal budget—a critical advantage for thin‑and‑light gaming laptops that also need to run AI‑enhanced software.
Graphics Integration: Likely Radeon RX 7600M XT or Better
The “370” suffix mirrors the naming of Radeon RX 6700 XT and the newer RX 7600 M XT series. AMD’s recent “HX” chips have bundled a Vega‑based integrated GPU for low‑power models, but high‑end HX parts now feature a discrete Radeon GPU on the same package (APU‑style integration) or a tightly coupled chiplet solution.
Given the performance tier, the “Ryzen AI 9 HX 370” is expected to ship with a Radeon RX 7600M XT, offering:
- Up to 2,500 stream processors and a 128‑bit memory interface.
- Support for DirectX 12 Ultimate, Vulkan 1.3, and AMD’s FidelityFX Super Resolution (FSR) 3.
- Dynamic Boost technology that can shift power between the CPU, GPU, and NPU based on workload.
This combination would put the chip in direct competition with Intel’s 13th‑gen “vPro i9‑13980HX” and Nvidia’s “GeForce RTX 4060 Laptop” solutions, targeting gamers and creators who also need AI‑centric features.
Memory, Cache, and Power Envelope
High‑end mobile processors have traditionally used a 45 W‑65 W TDP envelope, with the “HX” line leaning toward the higher end to sustain performance. The “370” is likely to follow this pattern, offering:
- Dual‑channel DDR5‑5600 support, with a maximum capacity of 64 GB.
- 1 MB of L2 cache per core (total 8 MB) and a shared 32 MB L3 cache, mirroring the Ryzen 9 7945HX.
- Integrated LPDDR5X options for ultra‑thin designs that prioritize power efficiency.
AMD’s “Infinity Fabric” clock speeds are expected to run at 2.2 GHz or higher, ensuring low latency between the CPU, GPU, and NPU. This fabric bandwidth is critical for AI workloads that need to move data quickly between cores and the accelerator.
Real‑World Use Cases: Why the AI Component Matters
While raw CPU and GPU performance are still the headline numbers, the inclusion of an on‑chip NPU opens new possibilities for laptop users:
- Content Creation: AI‑driven up‑scaling of 4K video, real‑time background removal in video calls, and smart photo enhancements can be performed locally, reducing reliance on cloud services.
- Gaming: Features like AMD’s “AI‑SuperResolution” can upscale lower‑resolution frames to 4K with minimal performance hit, giving gamers smoother experiences on smaller screens.
- Productivity: Voice assistants, real‑time transcription, and language translation become more responsive when the inference happens on the NPU rather than being sent to a server.
- Security: On‑device facial recognition and anomaly detection can run continuously without compromising battery life.
These scenarios are already being demonstrated on existing Ryzen AI laptops, and the “370” would simply raise the performance ceiling, making AI‑enhanced workflows more fluid.
What to Expect From OEMs and the Market
AMD’s strategy with the “AI” moniker has been to give OEMs a clear value proposition: a single silicon solution that covers CPU, GPU, and AI compute. Early adopters such as ASUS, Lenovo, and Dell have already announced “Ryzen AI” laptops equipped with the 7840U. If the “Ryzen AI 9 HX 370” materializes, we can anticipate:
- Gaming‑oriented laptops with 165 Hz or higher refresh‑rate panels, leveraging the AI up‑scaling engine.
- Creator‑focused machines with color‑accurate 4K displays and built‑in AI video‑editing tools.
- Business ultrabooks that advertise “on‑device AI security” as a selling point.
Pricing is likely to sit in the $1,600–$2,200 range for premium configurations, based on the cost of comparable Ryzen 9 HX chips and the added expense of the NPU.
Bottom Line: A Glimpse of the Future of Mobile Computing
If the “Ryzen AI 9 HX 370” lives up to the expectations set by AMD’s recent roadmap, it will represent a convergence of three traditionally separate performance domains: raw CPU power, high‑end graphics, and dedicated AI acceleration. This integration could simplify laptop designs, reduce power consumption, and enable new software experiences that are currently only possible in desktop or cloud environments.
Until AMD makes an official announcement, the specs discussed here remain informed speculation based on publicly available data and the company’s established naming conventions. Nonetheless, the trajectory is clear: AMD is positioning its mobile lineup not just as a place to play games or run office apps, but as a platform where AI becomes a first‑class citizen, right alongside the CPU and GPU.
