← All hardware
A purpose-built one-chip ML SoC for the full edge AI pipeline.
Pros
- Complete edge ML pipeline on one low-power chip
- Strong efficiency for embedded smart-vision
- Designed for harsh, power-constrained deployments
Cons
- Enterprise/OEM focus, less hobbyist-accessible
- Pricing typically via sales/volume, not public retail
- Smaller ecosystem than NVIDIA or Hailo
✓ Where it shines / best for
- Drones and robots needing self-contained vision + control
- Industrial and transportation smart-vision systems
- OEMs building power-constrained embedded AI products
✕ Not the best fit for
- Hobbyists wanting cheap off-the-shelf boards
- Datacenter-scale inference
- Buyers needing transparent public pricing
Features
- ✓ Single-chip MLSoC: ML accelerator + CPU + CV processor
- ✓ ~50 INT8 TOPS at ~5W (10 TOPS/W)
- ✓ Quad-core Arm Cortex-A65 CPU
- ✓ Synopsys EV74 computer-vision processor
- ✓ Handles full sensor-to-inference-to-control pipeline
- ✓ Palette software for model deployment
- ✓ Modalix next-gen line adds generative AI support
- ✓ Targets drones, robotics, transportation, healthcare
Pricing
| Plan | Price | Billing | Notes |
|---|---|---|---|
| MLSoC Developer Kit | Contact sales | one-time | OEM/volume pricing; not publicly listed at retail |
Pricing verified from the official source. Prices change often — confirm on the vendor's site before buying.
Specifications
| power | ~5W (10 TOPS/W) |
| memory | LPDDR (configurable) |
| form_factor | SoC / developer kit / modules |
| architecture | MLSoC: ML accelerator + quad-core Arm Cortex-A65 + Synopsys EV74 CVP |
| ai_performance | ~50 INT8 TOPS |
Sponsored
A full review is being generated for this product and will appear here shortly.