What shipped on Runpod in Q3 2026

Runpod’s recent releases focus on reducing operational complexity as workloads scale across inference, storage, training, APIs, and enterprise administration. Batch Jobs, currently in public beta, allow users to submit up to thousands of offline inference requests to Serverless endpoints and monitor them collectively, while worker affinity helps preserve GPU-resident KV caches by routing follow-up requests to the same worker when possible. Serverless endpoint creation now supports multiple starting sources, and Nano Banana Edit users must migrate to Nano Banana 2 Edit following the original model’s retirement. Global Volumes provide datacenter-independent, object-storage-backed shared storage suited to frequently read, infrequently written assets, while private AWS ECR integration enables secure image pulls without manually renewing short-lived registry credentials. Training clusters can now be expanded without rebuilding them, though workloads may need to be resubmitted to recognize the changed topology. REST API v2 unifies platform management features and introduces new metrics and visibility capabilities, while users of API v1 and GraphQL are advised to migrate before their scheduled deprecations. Runpod Enterprise has added Organizations for centralized billing, access management, SSO/SAML, auditing, tagging, and cost allocation, alongside the company’s ISO/IEC 27001:2022 certification.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.