Back to Gap Map Infrastructure / Edge AI hardware

Axelera Metis AIPU

documented / Overall score: 3.5

Edge AI inference accelerator from Axelera AI in the Netherlands, built on a quad-core digital in-memory-compute architecture. Each Metis chip delivers up to 214 TOPS at about 15 TOPS per watt and is sold as M.2 modules, PCIe cards and single-board computers. The Voyager SDK supports more than 100 models.

Openness

3 medium confidence

3.0

schematics: none
toolchain: open (Voyager SDK on public GitHub)
datasheets: brief (product pages/briefs public)
blobs: required (proprietary AIPU silicon + firmware)
retail: open_market (webstore + partners)
form_factor: module (sold as M.2 modules and PCIe cards)

Proprietary in-memory-compute silicon with a publicly available Voyager SDK on GitHub and retail purchasing, but board design files are not open.

https://github.com/axelera-ai-hub/voyager-sdk recorded 2026-08-13
public voyager-sdk repository under Axelera's own org, "This is a production-ready release of Voyager SDK" at v1.8, with release notes and a compatibility matrix in-tree
https://www.axelera.ai/ recorded 2026-08-13
Metis AIPU product pages and a Shop, "D-IMC technology performs matrix-vector multiplications in-place, delivering 50+ TOPs per core" and "At 15 TOPs/W" - marketing briefs and a store, with no board design files offered anywhere in the document set

Adoption

3 medium confidence

3.0

Commercially available via webstore and partners with a public, versioned SDK, but a younger ecosystem than Hailo/Coral. The Voyager SDK is at v1.8 with release notes and a release-compatibility matrix in-tree, and Axelera announces an ASRock Industrial collaboration integrating Metis AIPU cards into edge platforms.

https://github.com/axelera-ai-hub/voyager-sdk recorded 2026-08-13
v1.8 production release with RELEASE_NOTES.md and RELEASE_COMPATIBILITY_MATRIX.md in-tree
https://www.axelera.ai/ recorded 2026-08-13
a Shop link and an announcement that "ASRock Industrial ... integrates Axelera Metis AIPU cards into ASRock Industrial's edge AI platforms"

Capability

4 medium confidence

4.0

Each Metis chip is rated at up to 214 TOPS, and Axelera now lists language-model inference up to 8B parameters alongside its vision models. The Voyager SDK still labels its language-model tutorial experimental.

https://www.axelera.ai/ recorded 2026-08-13
"delivering 50+ TOPs per core at FP32-equivalent accuracy with zero data movement", "At 15 TOPs/W", and "100+ supported models and a single toolchain" for Voyager

Unchanged since 2026-09-29 (last edited, not re-checked)