Syntiant’s NDP120 is a low-power edge-AI system-on-chip designed to keep several neural networks running locally in an always-on device. In a January 6, 2021 report, EE Times described Syntiant’s target as multiple always-on networks within an under-1-mW power budget. That figure is a company-reported design claim, not a promise for every model or operating condition.
What is the Syntiant NDP120?
The NDP120 is a second-generation neural decision processor for battery-powered edge devices. Rather than sending every sound or sensor reading to a remote service, it is designed to process inputs on the device and make recognition or classification decisions locally.
Its main components divide the work:
- Syntiant Core 2: the neural accelerator that runs neural-network inference.
- Tensilica HiFi3 DSP: a programmable digital signal processor for audio feature extraction and front-end processing.
- Arm Cortex-M0: the controller that manages the system.
This is a chip-level design. It is distinct from a development board that incorporates the chip, such as the Arduino Nicla Voice.
How can it run several AI models at once?
Core 2 handles neural-network inference
Syntiant Core 2 uses near-memory computing: neural processing is closely coupled to on-chip SRAM, reducing the need to move data back and forth between separate processing and memory resources. EE Times reported that Core 2 supports convolutional networks, recurrent networks, LSTMs and fully connected networks. The reported quantization options are 1, 2, 4 and 8 bits, with a 16-bit inference mode also supported.
#1 Best Overall
- 2.4GHz Dual Mode WiFi + Bluetooth Development Board
- Support LWIP protocol, Freertos;ESP32 is a safe, reliable, and scalable to a variety of applications
- SupportThree Modes: AP, STA, and AP+STA
- Ultra-Low power consumption, Compatible with Arduino IDE
- 1PCS 30Pin ESP32 Development Board 2.4GHz WiFi Dual Cores Microcontroller Integrated with Antenna RF Low Noise Amplifiers Filters
Syntiant said Core 2 delivers 25 times the tensor throughput of its first-generation core and can support networks of up to 7 million parameters. These are company figures reported by EE Times in 2021, not independent benchmark results. Parameter capacity alone does not establish the accuracy, latency or power use of a particular model.
The DSP and accelerator can serve different parts of an audio pipeline
The HiFi3 DSP can handle programmable audio processing while Core 2 runs neural networks. That division makes it possible to combine front-end tasks with recognition models instead of treating voice-command recognition as the only workload.
Rank #2
- Dual-core processor: The ESP32 module is based on the powerful ESP32-S3-WROOM N16R8 module and is equipped with a dual-core 32-bit LX7 processor. Its excellent AI computing performance, real-time processing capabilities, and low power consumption make it ideal for image recognition, edge AI, and complex IoT applications
- Integrated 2-megapixel OV3660 camera: Built-in OV3660 camera to capture clear images and stream video in real time. Perfect for smart surveillance, face recognition, and AI-based computer vision projects. It is the preferred solution for DIY makers and professionals to build camera-enabled IoT systems
- Dual Type-C ports for OTG and serial debugging: Designed with two USB Type-C interfaces - one supports USB OTG for host/device functions, and the other provides TTL serial for easy programming and debugging
- Shared antenna: Supports IEEE 802.11b/g/n Wi-Fi (2.4GHz) and Bluetooth 5 (LE and Mesh), using shared antennas to optimize wireless performance. Enhanced 2 Mbps PHY and long-distance communication (Coded PHY) ensure stable multitasking in harsh environments
- Multi-scenario applications: The ESP32 S3 development board maintains high stability even at high temperatures, making it ideal for industrial environments, educational purposes, and AI-driven projects. It is a versatile choice for robots, smart devices, and machine vision in lab or field applications
The practical examples in the 2021 report include echo cancellation, beamforming, noise suppression, speech enhancement and speaker identification running alongside voice-command recognition. Syntiant’s NDP120 brief also describes near-field and close-talk interfaces, multiple wake words and local commands, acoustic-event and scene classification, and multi-sensor fusion. These describe intended capabilities; they do not mean every listed task is guaranteed to run together at a particular power level.
What does “under 1 mW” mean?
EE Times presented under 1 mW as the NDP120’s power target for multiple always-on neural networks in edge devices. It is not a universal consumption figure for any combination of models. The reported figure is not accompanied here by a specific model set, measurement method, or full operating conditions, so it cannot be used to predict a device’s exact power draw or battery life.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteRank #3
- 【ESP32-C3 RISC-V Development Board】 Built with the ESP32-C3 32-bit RISC-V chip (160MHz), featuring Arduino/CircuitPython support and multiple development ports. Ideal for IoT and edge AI projects.
- 【Outstanding RF & Long-Range Connectivity】 Equipped with U.FL antenna for stable Wi-Fi/BLE5.0 communication over 100m. Complete RF performance ensures reliable IoT connectivity.
- 【Ultra-Low Power & Battery-Friendly】 4 working modes, including deep sleep at 44μA. Onboard battery charge IC supports Li-ion/LiPo, perfect for wearables and wireless IoT.
- 【Thumb-Sized & Production-Ready】 Compact 21x17.5mm design with SMD/Breadboard-friendly layout. Single-sided component mounting ensures sleek integration into wearables.
- 【Rich I/O & Edge Computing】 11 digital I/O (PWM) + 4 analog I/O (ADC), plus UART/IIC/SPI/IIS ports. Optimized for TinyML and edge AI applications.
The distinction matters because actual consumption depends on the workload and how the chip is operating. Syntiant CEO Kurt Busch described the product goal in the January 6, 2021 EE Times interview: “The NDP120 can bring the level of performance that you would typically find in a plugged-in smart speaker to a battery powered device, that’s really the goal for this product.” This is a statement of intent from the company’s CEO, not a measured comparison with a particular smart speaker.
How does an NDP120 compare with an MCU-plus-NPU or cloud pipeline?
The NDP120 is one example of local, integrated inference. The alternatives below are broad architectural categories, not comparisons against named products: exact power, model capacity and latency depend on the specific hardware, software and workload.
Rank #4
- 【ACEBOTT ESP32 Development Board】 - Powerful WiFi and wireless development board, driven by the rugged ESP 32 module, seamlessly integrated with Arduino IDE. With Hall sensors, high-speed SDIO/SPI, UART, I2S and I2C, it is the cornerstone of IoT and smart home innovation.
- 【Wi-Fi/Bluetooth and Arduino Cloud Compatibility】 - This board uses 2.4GHz dual-mode WiFi and wireless chips with low-power technology, which are RoHS-compliant, simplifying wireless communication and allowing you to easily connect devices and platforms. Whether you are using a compatible Arduino IDE or exploring other development environments, our board can easily adapt to your needs.
- 【Improved and Professional Edition】 - All IO pins are brought out for easy development; no additional breadboard is required; the Type-C interface is equipped with electrostatic discharge protection diodes and transient voltage suppression diodes to protect the chip from damage by electrostatic breakdown and various surge pulses. In addition, it is equipped with a freeRTOS operating system, which is very suitable for the Internet of Things, smart homes, and building smart robots/game consoles.
- 【Easy to Use】- The ACEBOTT ESP-32 Development Board includes everything you need to support the microcontroller. Just connect it to a computer via a USB cable or use an AC-DC adapter or battery to power it to start using it. Whether you are an experienced developer or a hobbyist, this development board can provide you with the tools you need for unlimited innovation.
- 【 Install Plugins And Download Drivers】: This ESP32 development board includes detailed instructions on how to download plugins and all necessary programs and codes from the network environment. The path is: ACEBOTT official website - Resources - WIKI.
| Decision factor | NDP120 | Conventional MCU plus NPU | Cloud-based processing |
|---|---|---|---|
| Power budget | Syntiant reported an under-1-mW target for multiple always-on neural networks; workload and operating conditions are not specified here. | Not stated for any specific system; depends on the MCU, NPU and workload. | Not stated for any specific service or device; the end-to-end system also depends on its connection and local hardware. |
| Model capacity | Up to 7 million parameters, according to Syntiant as reported by EE Times in 2021. | Not stated; depends on the selected components and software. | Not stated; depends on the service and deployed model. |
| Concurrent workloads | Designed for multiple networks; reported examples combine audio front-end tasks with voice commands. No guaranteed concurrent workload count is established here. | Depends on the particular NPU, MCU and software; no specific count is stated. | Depends on the service and workload; no specific count is stated. |
| Audio and sensor I/O | Supports up to seven audio streams, according to the 2021 report, and Syntiant describes multi-sensor fusion. The report does not establish performance for every seven-stream workload. | Depends on the selected components and interfaces. | Depends on the device’s capture hardware and connection to the service. |
| Latency and privacy | Local inference can avoid sending each input to a cloud service, which can reduce dependence on a network round trip and keep processing on-device. Actual latency and privacy behavior depend on the product implementation. | Local inference can offer the same general architectural advantages; exact results depend on implementation. | Processing requires sending inputs to a service, so the connection and service affect latency and data handling. |
| Development compatibility | The specific toolchain compatibility is not stated in the information reported here. | Depends on the chosen vendor’s development tools and model support. | Depends on the cloud service’s APIs and deployment requirements. |
The NDP120’s defining trade-off is integration: audio processing, system control and neural inference are combined in a chip aimed at local, always-on use. A cloud pipeline may suit workloads that rely on remote services, while a separate MCU-plus-NPU design can vary widely with component choice. The figures in the NDP120 column remain Syntiant-reported capabilities, not an independent head-to-head test.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Can you prototype with an NDP120?
Syntiant’s hardware portfolio identifies the Arduino Nicla Voice as a platform developed with Arduino and powered by NDP120. It is a physical board for prototyping always-on speech recognition and concurrent AI models; it is not the NDP120 chip itself. Syntiant’s portfolio labels NDP120 as being in mass production. That portfolio status does not establish a particular distributor’s stock, price, lifecycle status or regional availability.
Best Value
![Arduino® UNO™ Q 2GB[ABX00162] - Hybrid Board, Qualcomm Dragonwing QRB2210 microprocessor (MPU) & STM32U585 Microcontroller(MCU), AI Vision, Voice, IoT, Robotics, Linux Debian OS, Wi-Fi 5, USB-C](https://m.media-amazon.com/images/I/51cl4xygwGL._SL160_.jpg)
- Dual-Brain Hybrid Power: Combines the Qualcomm Dragonwing QRB2210 MPU (Quad-core Arm Cortex-A53 @ 2.0 GHz CPU, Adreno GPU, AI acceleration) and the real-time, low-power STM32U585 MCU for advanced applications like object recognition, voice commands, and motion detection.
- AI & Linux Capabilities: Unlocks AI-powered vision and sound solutions; runs Linux Debian OS for coding in Python and supports the Arduino ecosystem with libraries and Sketches; quick start with Arduino App Lab.
- Advanced Features: Equipped with 2 GB LPDDR4 RAM, 16 GB eMMC built-in storage, ideal to develop in PC-connected mode, running the OS, Python scripts, and basic network services (SSH) without a demanding GUI or heavy multitasking; great for lightweight AI and memory-optimized TinyML applications, needing local storage for basic OS and core libraries. Dual-band Wi-Fi 5 (2.4/5 GHz), Bluetooth 5.1, and high-speed headers for vision, audio, and display peripherals.
- Seamless Expansion & Connectivity: Features the classic UNO form factor for shields compatibility, an 8x13 LED matrix, and a Qwiic connector for easy expansion with Modulino nodes; power and connect via the USB-C connector.
- Intended Use & Development: The perfect platform for prototyping robotics or IoT projects, empowering innovators with a unified development experience to mix Arduino Sketches, Python scripts, and containerized AI models in a single interface.
At the NDP120’s 2021 launch, EE Times said the chip was sampling and expected to ship in production volumes in summer 2021. That is historical launch timing, separate from the company’s later mass-production label.
What the performance figures do—and do not—show
The 25× throughput, 7-million-parameter capacity and under-1-mW target are Syntiant claims reported by EE Times on January 6, 2021. No independent published test report is established here to verify those figures under a defined benchmark or representative workload. They describe the company’s stated design capabilities, not a guaranteed result for every application.
Busch characterized Core 2 as the result of a lengthy development effort in the same interview: “The Syntiant Core 2 takes about three years of learning to build a very flexible core that can scale up to much larger applications.” The statement gives context for the company’s design ambitions; it does not substitute for workload-specific measurements.
Syntiant later introduced NDP115 in 2023 and NDP250/Core 3 in 2024. Those are later products, and their specifications should not be attributed to the NDP120. In particular, the company’s description of NDP250 as a 30-GOPS, five-times-throughput product applies to that later product, not Core 2.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
