EverOS v1.5.1 is here. A small open-source update for a frustrating problem: memory searches queuing up under load. We increased the default query-embedding concurrency from 4 to 16, giving hybrid and vector search more room to handle concurrent requests. In our five-minute test with 10,020 indexed entries at 10 requests per second: • p95 latency dropped from 101s to 1.45s • All 3,000 requests were served No migration needed, just upgrade. If your embedding endpoint has tight rate limits, tune the concurrency to match. Prefer to build without managing the infrastructure? EverOS Cloud brings managed memory to your existing agent workflows. Save useful context after a task, retrieve it before the next, and give your agents access to user preferences, business knowledge, and past experience through APIs. Start with one recurring workflow and let useful context accumulate across runs. Subscribe to EverOS Cloud Pro now and get 3× the standard monthly usage for your first 3 months. Cloud: