Efinix Releases TinyML Platform for Highly Accelerated AI Workloads on Its Efficient FPGAs
Quad-Core, Linux-Capable RISC-V Uses Custom Instruction Capability to Achieve AI Acceleration
SANTA CLARA, Calif.—Nov 16, 2022—Efinix®, an innovator in programmable product platforms and technology, today announced the release of its TinyML Platform for artificial intelligence (AI) acceleration on its innovative range of FPGAs. The TinyML Platform works with the high performance and Linux-capable Sapphire RISC-V processor. Comprising a model profiler and graphical user interface for accelerator selection, the TinyML Platform delivers an accelerated AI capability to the quad-core-capable Sapphire core.
“We are seeing an increasing trend to drive AI workloads to the far edge where they have immediate access to raw data in an environment where it is still contextually relevant. Providing sufficient compute for these AI algorithms in power and space constrained environments is a huge challenge,” said Mark Oliver, Efinix VP of Marketing. “Our TinyML Platform harnesses the potential of our high performance, embedded RISC-V core combined with the efficiency of the Efinix FPGA architecture and delivers them intuitively to the designer, speeding time to market and lowering the barrier to AI adoption at the edge.”
The TinyML Platform leverages the RISC-V custom instruction capability to deliver a scalable library of accelerated instructions customized for the requirements of the chosen TensorFlow Lite model. Instructions vary in complexity and size and can be selected through a graphical user interface to deliver the resource / acceleration tradeoff needed for the end user’s application. Time to market is further accelerated for the designer by leveraging the Edge Vision SoC Framework as the starting point for AI model implementations. Both the Edge Vision SoC Framework and the TinyML Platform are available to the open-source community on the Efinix GitHub page.
About Efinix
Efinix®, an innovator in programmable products, is committed to delivering the low power and reconfigurability of its high-performance Titanium FPGA silicon platforms for applications in the mainstream market. Efinix FPGAs deliver power, performance, and area advantages over traditional FPGA technologies, unlocking new applications and delivering rapid time to market. Titanium devices range from 35K to 1M logic elements, have a small form-factor, low-power, and are priced for high-volume production. The Efinity® Integrated Development Environment provides a complete FPGA design suite from RTL to bitstream.
|
Related News
- Efinix Releases Topaz Line of FPGAs, Delivering High Performance and Low Power to Mass Market Applications
- Omnitek Releases Highly Optimised 3D LUT IP for FPGAs
- Achronix Releases Groundbreaking Speedster AC7t800 Mid-Range FPGA, Driving Innovation in AI/ML, 5G/6G and Data Center Applications
- RaiderChip launches its Generative AI hardware accelerator for LLM models on low-cost FPGAs
- Efinix Rolls Out Line of FPGAs to Accelerate and Adapt Automotive Designs and Applications
Breaking News
- Breker RISC-V SystemVIP Deployed across 15 Commercial RISC-V Projects for Advanced Core and SoC Verification
- Veriest Solutions Strengthens North American Presence at DVCon US 2025
- Intel in advanced talks to sell Altera to Silverlake
- Logic Fruit Technologies to Showcase Innovations at Embedded World Europe 2025
- S2C Teams Up with Arm, Xylon, and ZC Technology to Drive Software-Defined Vehicle Evolution
Most Popular
- Intel in advanced talks to sell Altera to Silverlake
- Arteris Revolutionizes Semiconductor Design with FlexGen - Smart Network-on-Chip IP Delivering Unprecedented Productivity Improvements and Quality of Results
- RaiderChip NPU for LLM at the Edge supports DeepSeek-R1 reasoning models
- YorChip announces Low latency 100G ULTRA Ethernet ready MAC/PCS IP for Edge AI
- AccelerComm® announces 5G NR NTN Physical Layer Solution that delivers over 6Gbps, 128 beams and 4,096 user connections per chipset
![]() |
E-mail This Article | ![]() |
![]() |
Printer-Friendly Page |