main
Message drafts (#3044)
Revert "Cleanup `handle_worker()` in preparation for #2815 (Stop generation early) (#3573)" (#3644)
Initial implementation of the inference system (#869)
added async postgres to inference (#1961)
Initial inference documentation pass (#3330)
Add rate limits to assistant chat messages (#3514)
set PROMETHEUS_MULTIPROC_DIR env var (#2956)