Client no longer retries when the server rate-limits the request
Client no longer retries when the server rate-limits the request: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). We have a pymilvus client that issues a fairly high rate of insert / search calls against a cluster that has quota / rate-limit enabled on the server side. On…
The task
We have a pymilvus client that issues a fairly high rate of insert / search calls against a cluster that has quota / rate-limit enabled on the server side. On older versions, when we hit the quota, the client would back off and retry transparently — our application barely…
Part of XiaomiMiMo/MiMo-V2.6-RL-oss.