Near-Memory AI Acceleration Platform Delivering Maximum Performance per Watt
INTERA provides a family of ultra-low power AI accelerators, from low-latency RISC-V custom instructions to near-memory computing engines, enabling TinyML systems to achieve maximum performance-per-watt with minimal silicon and memory footprint.
INTERA Cognit IP Selection Guide
| AI Accelerator | Architecture | Operation accelerated | Compute size (*PE) and scalability |
|---|---|---|---|
| Falcon | Monolithic, streaming processor | MxV | Single PE, 2 to 32-bit implementation |
| Hawk Advance | RISC-V custom instruction | MxV / MxM⁽²⁾ | Single PE, 2 to 32-bit implementation. Entry-level version, scales to Hawk Pro |
| Hawk Pro | RISC-V custom instruction | MxV / MxM⁽²⁾ | 1, 2, or 4 PEs. Configurable 2 to 32-bit |
| Eagle | Near-memory computing | MxV⁽²⁾ | 4, 16 PEs; tested up to 32 banks/PEs. Configurable 2 to 32-bits |
| Harrier | Monolithic, streaming processor | MxM | Scalable matrix array; 8x8 = 64 PEs; configurable as 32x32, 32x64, 64x64, … |
Software Suite for Classical AI
(CNN, LSTM, MLP/FCNN)
INTERA provides an open-source tool suite which seamlessly interoperates with existing deep learning development frameworks, enabling fast and efficient deployment of AI models on INTERA accelerators.
AI accelerator is a Edge AI Accelerator IP core from Intera Group listed on Semi IP Hub.
Engineers should review the overview, key features, supported foundries and nodes, maturity, deliverables, and provider information before shortlisting this Edge AI Accelerator IP.
Yes. Buyers can compare this product with similar semiconductor IP cores or IP families based on category, provider, process options, and structured technical specifications.