Launched an Inference Engine bundle claiming up to 67% lower production AI inference costs.

This is a direct category move targeting developers and small teams who are being priced out of AI inference by AWS/Azure/GCP — DigitalOcean's core constituency. Offering serverless, batch, and dedicated inference in one bundle reframes DO as a serious AI infrastructure provider, not just a cheap VPS host. Competitors in DevTools and AI infra need to respond on price and flexibility, and buyers evaluating inference hosting now have a credible SMB-friendly alternative to benchmark against.