Series A · Hardware
Wafer raises $40M Series A to optimize AI inference workloads
Wafer benchmarks models, serving software and hardware together, then keeps optimizing against live traffic.
Continuous optimization of model serving
Wafer provides an inference endpoint for AI teams and enterprises. Customers bring a model, request pattern and service targets; the platform evaluates batching, caching, routing, decoding kernels and available hardware to choose a serving setup that meets those requirements while preserving output correctness.
After deployment, Wafer observes real traffic and continues tuning performance and reliability. DigitalOcean reports using the platform to improve inference speed for a model running on AMD hardware.
Use of funds
Procure high-performance compute and wafer-level manufacturing
Expand engineering
Accelerate customer onboarding
Funding history
Series A · This round$40M