E1M-X · variant
Coming soonE1M-X V2H+M1
V2H plus DX-M1 reaches 33 TOPS dense (105 TOPS sparse) for high-throughput inference at the edge.
Sampling Q3 2026. Target specs locked; final benchmarks pending.
In partnership with
§ What you'll get at launch
· Full specs + benchmarks
Final compute, memory, and interface numbers — measured on production silicon, not target estimates.
· Datasheet + HRM
Production-grade datasheet, hardware reference manual, and pinout JSON. One-click download.
· Order & availability
SKU, package, MOQ, lead time, and distributor links published when sampling opens.
· Reference designs + SDK examples
Working examples on the unified Alp SDK. Same C/C++ API as the shipping variants.
§ Where it fits in the lineup
AI compute ladder · log scale · TOPS
range
0.25–0.5 TOPS→33 TOPS
- 0.25–0.5TOPSE1M-AEN35 × 35 mm
- 0.5TOPSE1M-N93Coming soon35 × 35 mm
- 4TOPSE1M-X V2N45 × 65 mm
- 8TOPSE1M-X V2HComing soon45 × 65 mm
- 29TOPSE1M-X V2N+M145 × 65 mm
- 33TOPSE1M-X V2H+M1Coming soon45 × 65 mm
always-on sensingedge inferencereal-time + linux