1. Technology overview
Axelera AI designs AI accelerator chips (AIPUs) based on in-memory computing, in which computation happens inside SRAM. Its products include the Metis AIPU, the next-generation Europa AIPU and the announced Titania chiplet, offered as M.2 modules, PCIe cards and systems for embedded, edge and server deployment. The Voyager SDK compiles, quantizes and runs AI pipelines described in YAML, with a model zoo. Target applications include industrial manufacturing, retail, security, healthcare, smart cities, robotics, computer vision and on-premise LLM inference.
2. Highlighted features & specifications
- In-memory computing AI accelerators (Metis, Europa AIPUs)
- Embedded, edge and server form factors (M.2, PCIe cards, systems)
- Voyager SDK: compile, quantize and run YAML-described pipelines
- Post-training quantization without retraining
- On-device LLM inference (up to 8B parameters on Metis, 32B on Europa per vendor)
One platform connecting startups, corporations, research institutions, regions and investors











Playbook & Handbook · February 6, 2026
Winning Accelerator Playbook 2026 by BambuUP
EDTECH: LEVERAGING AI-POWERED TECHNOLOGIES TO TRANSFORM THE LEARNING EXPERIENCE








































