Straightforward pricing for fab teams

Annual billing available with a 15% discount on all tiers. All plans include on-premise deployment and SECS/GEM tool integration.

Starter

$4,800/mo

Single fab, up to 3 inspection tool connections. For teams piloting AI-assisted defect review on one process module.

  • Up to 3 inspection tool integrations
  • CDSEM and optical review classification
  • Pre-trained defect library (47 types)
  • Single-layer defect correlation
  • Email alerts to up to 5 engineers
  • On-premise deployment
  • Standard support (next business day)
Get Started

Enterprise

Custom

Multi-fab, multi-site deployments. ITAR-compliant environments, customer-managed encryption, dedicated model development.

  • Unlimited tool integrations
  • Multi-fab, multi-site deployment
  • Air-gapped network support
  • Customer-managed encryption keys
  • Custom defect type training
  • Dedicated yield engineer support
  • SLA: 99.7% uptime, 2-hour response
  • On-site integration assistance
Contact Sales

All prices in USD. Annual billing saves 15%. Volume discounts available for multi-fab contracts.

Common questions about pricing and deployment

Is there a free trial available?
The Growth tier includes a 30-day proof-of-concept period using your existing inspection data. During the trial, we configure tool integrations and deliver a baseline yield impact assessment. If you don't see a meaningful defect detection improvement within the trial period, you owe nothing.
How do you count "tool integrations"?
Each physical inspection tool — regardless of how many chambers it has — counts as one integration. A KLA Surfscan SP7 with four load ports counts as one. If you have a cluster tool where the inspection module is one of several process chambers, only the inspection unit counts.
Can we switch between plans as our deployment grows?
Yes. Upgrades take effect immediately and are prorated for the remainder of your billing period. Downgrades take effect at the next renewal date. Most customers start on Starter for a single process module and upgrade to Growth once they validate the yield improvement across the full line.
How does SynthKernel handle our wafer image data security?
All model inference runs on-premise. Wafer images, die maps, and process parameters never leave your facility network. The platform does not require outbound internet access for normal operation. Audit logs of every inference event are stored locally and are available for your security team's review at any time.
What hardware does the platform require?
The Starter and Growth tiers run on a single on-premise server with an NVIDIA A30 GPU or equivalent (24GB VRAM minimum). Most fabs already have equivalent hardware in their MES server room. Enterprise deployments with high-frequency inspection loads may require additional GPU nodes, which we size during the pre-sales assessment.
Does the platform require access to our MES?
MES integration is optional. SynthKernel can run on inspection tool output alone. However, connecting to your MES enables process history lineage tracking and lot disposition workflows, which significantly improve root-cause analysis quality. Most customers add MES integration in the second month of deployment.

Not sure which plan fits your fab?

Talk to our team. We'll assess your inspection toolset and process complexity, and recommend the right starting point — no sales pressure.