EnCharge AI announced the EN100 in June 2025 as an analog AI accelerator for laptops, workstations and other client devices. Its central idea is to perform some calculations inside a memory array using charge stored on capacitors, reducing the need to move data between separate memory and compute units. EnCharge has reported high performance and efficiency figures, but the cited announcement does not establish independent benchmark results or broad consumer availability.
What is the EnCharge EN100?
The EN100 is EnCharge AI’s first announced product: an accelerator intended to run AI workloads on client devices rather than relying on a remote data center for every operation. In its June 13, 2025 report, EE Times described the chip as targeting laptops, workstations and other client platforms.
EnCharge’s initial product concepts included two distinct card configurations, not one combined design:
| Configuration | Reported specifications |
|---|---|
| Single-chip M.2 card | 32 GB LPDDR and an 8.25 W envelope, according to EnCharge as reported by EE Times. |
| Four-chip PCIe card | Up to 1 POPS INT8, 128 GB LPDDR and a 40 W envelope, according to EnCharge as reported by EE Times. |
The M.2 label describes the card’s form factor; it does not by itself establish that the card will work in a standard laptop M.2 slot. Host compatibility, system integration and product availability are separate questions.
#1 Best Overall
- ✅Powered by 26 Tera-Operations Per Second (TOPS) Hailo-8 AI Processor. 2.5W typical power consumption
- ✅Scalable, enabling simultaneous processing of multi-streams & multi-models
- ✅Enabling real-time, low latency and high-efficiency AI inferencing on the edge devices
- ✅Supports TensorFlow, TensorFlow Lite, ONNX, Keras, Pytorch frameworks
- ✅Supports Linux and Windows. Supports the temperature range of -40°C to 85°C
How does its analog compute-in-memory design work?
In a conventional system, processors repeatedly fetch model weights from memory and move data to the compute units. Those transfers can consume energy as well as time. Compute-in-memory designs seek to perform some operations where the weights are stored, limiting that movement.
EE Times reported that EnCharge’s approach uses charge stored on capacitors in a memory array for computation. CEO Naveen Verma described the capacitors as metal structures made from interconnect layers found in standard foundry processes. EnCharge’s technology page describes its broader approach as charge-domain computation using metal capacitors and presents it as a way to address signal-to-noise limits. These are explanations of the company’s architecture and rationale; they do not, by themselves, prove a measured advantage over other accelerators.
Rank #2
- High-Performance AI Processing: The MX3 is designed to handle the most demanding AI computer vision workloads, delivering exceptional performance and efficiency.
- Flexible Integration: The MX3 can be easily integrated into your existing systems via its M.2 M-key form factor and support for Linux operating systems.
- Energy Efficient: The MX3 is designed to provide high performance while minimizing power consumption.
- Comprehensive Software Development Kit (SDK): The MX3 is supported by a comprehensive SDK that simplifies development and deployment.
- Hardware compatability: The MX3 is compatible with the PCI-SIG M.2 M-key 2280 Specification. It can be used with the Raspberry Pi 5 with a M-key 2280 HAT.
Verma summarized the design challenge to EE Times as two-sided: AI needs efficient math, but it also moves large amounts of data. He said in-memory computing has the potential to address both concerns. That is the architectural motivation for EN100, not a universal result established across workloads.
Why is the accelerator a hybrid rather than fully analog?
EnCharge’s reported design combines analog and digital engines. The analog accelerator handles 8-bit and 4-bit precision work; on-chip digital engines handle higher-precision and floating-point operations. A compiler maps workloads across the available engines.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →This division matters because an AI accelerator’s headline throughput alone does not describe every model’s performance. Precision, supported operations, software mapping and the workload all affect what can run efficiently. The compiler is part of the system, not an optional detail: the design depends on software placing work on the appropriate analog or digital engine.
What performance and power figures has EnCharge claimed?
EE Times reported EnCharge’s stated figure of 200 TOPS at INT8 and an efficiency claim of more than 40 TOPS/W. The report does not provide an independent head-to-head test against competing PC accelerators under common workloads and measurement conditions, so these figures should be read as company-reported specifications rather than independently verified comparative results.
Rank #4
- ✅Powered by 26 Tera-Operations Per Second (TOPS) Hailo-8 AI Processor. 2.5W typical power consumption
- ✅Scalable, enabling simultaneous processing of multi-streams & multi-models
- ✅Enabling real-time, low latency and high-efficiency AI inferencing on the edge devices
- ✅Supports TensorFlow, TensorFlow Lite, ONNX, Keras, Pytorch frameworks
- ✅Supports Linux and Windows. Supports the temperature range of -40°C to 85°C
The article also characterized Microsoft’s Copilot-enabled laptop acceleration threshold as 40 TOPS, citing Verma. That threshold is one reference point in the AI-PC market, not a complete measure of a laptop’s AI capability or a guarantee that a device meeting it will perform a particular task well.
Meaningful comparisons require matching the details behind the numbers: precision and workload, whether power refers to the accelerator or the entire system, memory capacity and bandwidth, supported models and operators, physical form factor and host compatibility, and whether the product is shipping or only sampling. The EN100 figures cited above do not settle those comparisons.
Best Value
- DEEPX DX-M1M NPU: Powered by the DEEPX DX-M1M neural processing unit, purpose-built for efficient on-device AI inference workloads.
- COMPACT M.2 2242 FORM FACTOR: Fits the standard M.2 2242 slot, making it easy to integrate into embedded systems, edge devices, and compact computing platforms.
- EDGE AI ACCELERATION: Designed to accelerate deep learning inference at the edge, enabling real-time AI applications without relying on cloud connectivity.
- RADXA AICORE MODULE: The Radxa AICore DX-M1M delivers a plug-and-play AI compute solution ideal for robotics, smart cameras, and industrial automation.
- WARRANTY AND ORIGIN: Backed by a 1-year manufacturer warranty and crafted with quality components for reliable long-term performance in demanding environments.
Why did EnCharge choose PCs first?
Verma told EE Times that the PC market offered a focused opportunity and a value proposition around energy efficiency. The company’s case for local AI also centers on personalized or specialized models that users may want to run within compliance or security constraints, while working within the power and space limits of client devices.
EE Times reported that EnCharge was engaging with laptop and client-platform OEMs, ODMs and software companies. The report did not name a laptop partner or establish that a specific system had shipped with EN100. The PC-first strategy therefore describes the company’s target market and reported partner discussions, not confirmed retail deployments.
Can you buy an EN100-equipped laptop now?
The June 13, 2025 EE Times article said strategic customers were expected to receive samples later that year. That plan does not confirm whether sampling occurred, whether the product has since entered broader production, or whether consumers can buy an EN100-equipped PC. The available sources do not identify a retail listing or a consumer upgrade path.
EnCharge’s contact page invites inquiries about technology and partnerships, but it is not a product listing. Until a specific system and compatibility information are announced, treat “EnCharge EN100 AI accelerator” as a product-discovery search rather than a purchase recommendation.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →
