AI article
EmbeddingGemma 2 on Cloud Run: an int8 CPU Service Against an NVIDIA L4
Community description: Step by step deployment of Google's EmbeddingGemma 2 text embeddings to Cloud Run, on a 1 vCPU instance with an int8 repack and on an NVIDIA L4, with tests, a semantic search demo, benchmarks and a cost per million documents.
Dev.to | Oct 13, 2026 | xbill
Automated excerpt
Cloud Run runs a container image as an HTTPS service. A GPU service is billed per instance: the L4 is charged for every second the instance is up. An L4 instance costs about ten times as much per hour as a 1 vCPU instance.
Selected automatically from source text; not independently written or fact-checked. Read the original for full context.