The E4, E6, and E8 Ensemble series enable Gen AI in Edge devices for the first time
15 MB +
Integrated Memory
3670 DMIPS +
Combined Compute Performance
450 GOPs +
AI Performance
The E4, E6 and E8 Ensemble MCUs and fusion processors feature an improved high-throughput memory architecture and an upgraded neural processing unit (NPU), the Ethos ™ -U85. This unlocks hardware acceleration for transformer networks enabling generative AI workloads at the Edge. Transformer networks enhance endpoint applications by quickly contextualizing new data with key insights from past inputs. For localized use this results in an instant, accurate results when using multimodal AI in endpoint products focused on vision, voice, text, and sensor fusion, without relying on the cloud.
The Gen AI enabled Ensemble devices are based on Alif’s proven-in-use system architecture
Alif’s E4, E6, and E8 series embrace the latest energy-efficient neural processing technology supporting transformer networks to enable generative AI in products at the edge
Efficient NPUs for Generative AI
Additional AI/ML Acceleration
Realtime Processing (RTOS, Bare Metal)
Application Processing (Linux, HLOS, RTOS)
Power Management
Security
SRAM
Non-Volatile Memory
Graphics Imaging
Memory Expansion
ISP Acceleration Unit and JPEG Encoder
Audio
Digital Interfaces
Analog Interfaces
Never Before Seen Scalability
The Ensemble family is an Arm-based microcontroller that scales from single core to a new class of multi-core devices, that combine up to two Cortex-M55 MCU cores, up to two Cortex-A32 microprocessor cores capable of running high-level operating systems, up to two Ethos-U55 NPUs for artificial intelligence machine learning acceleration, and one Ethos-U85 NPU supporting transformer networks and enabling generative AI at the Edge.
