A weekly newsletter of the best news, articles and projects about GraphQL

Seven GraphQL conferences just landed on confs.tech. Time to fight your calendar and travel budget.

Share how your team adopts, operates, and scales GraphQL Federation—including AI and LLM workloads—in the vendor-neutral State of Federation 2026 survey.

Tickets for GraphQLConf 2026 are live. May 6–7, back at Meta's Menlo Park campus.

GraphQL Workbench is a new VS Code extension for schema design with built-in AI, from the creator of Apollo Workbench. It runs entirely local and free using Ollama and OSS models, featuring natural language operation generation, local vector storage via PGLite, federation support, schema linting, and AI-powered design analysis. It can embed even massive schemas (GitHub's 8000+ types in ~2 minutes)

Cosmo Streams adds three router handlers to enable subscription authorization, initial state delivery, and per-subscriber event filtering without building custom infrastructure.

You can now configure a distributed cache (Redis, Memcached, etc.) as a L2 cache for persisted documents, sitting between the in-memory LRU cache and the CDN. This reduces CDN calls, shares cache state across instances, improves performance and resilience in serverless and multi-instance deployments.