vh c0d3384c68
feat(litellm): embedder primary/failover split — qwen3-embedding on esh-ml1, new qwen3-embedding-bulk on nh3-ml1 ...
Prime's ruling 2026-10-05 (relayed by albok-dev, confirmed directly) replaces
the 2026-09-26 simple-shuffle load-share for embeddings only; the reranker
groups are unchanged. Routing uses LiteLLM per-deployment `order` (semantics
read at source in v1.97.0): lowest order among healthy deployments, and a
failed request retries at the next order.
Verified after the restart (43 s gateway outage): 6/6 calls per alias landed
on its primary. With nh3-ml1 tei-embed stopped, qwen3-embedding-bulk served 6/6
from esh-ml1 with 200s; once nh3-ml1 was back, bulk returned 6/6 to nh3-ml1.
The albok-service key now covers qwen3-embedding-bulk (200); the Nemi key does
not (403).
2026-10-05 00:32:42 -07:00
2026-09-07 15:08:56 -07:00
2026-08-24 08:54:06 -07:00
2026-10-04 23:33:24 -07:00
2026-09-19 06:49:11 -07:00
2026-09-07 15:08:56 -07:00
2026-09-15 08:41:44 -07:00
2026-09-27 01:51:05 -07:00
2026-04-20 22:14:00 -07:00
2026-09-27 01:56:04 -07:00
2026-09-28 16:36:44 -07:00
2026-04-20 22:28:30 -07:00
2026-09-12 23:05:29 -07:00
2026-09-07 15:08:56 -07:00
2026-06-02 10:52:04 -07:00
2026-09-25 23:58:08 -07:00
2026-09-07 15:08:56 -07:00
2026-09-07 15:08:56 -07:00
2026-09-14 07:28:26 -07:00
2026-09-12 23:05:29 -07:00
2026-09-07 15:08:56 -07:00
2026-09-12 23:05:29 -07:00
2026-08-11 07:46:11 -07:00
2026-09-25 15:55:33 -07:00
2026-10-05 00:32:42 -07:00
2026-09-12 22:00:50 -07:00
2026-09-14 19:09:27 -07:00
2026-09-12 23:05:29 -07:00
2026-08-22 09:50:51 -07:00
2026-09-07 15:08:56 -07:00
2026-09-26 00:07:50 -07:00
2026-08-16 15:15:02 -07:00
2026-09-12 22:00:50 -07:00
2026-08-24 14:19:49 -07:00
2026-09-12 23:05:29 -07:00
2026-10-01 01:38:42 -07:00
2026-09-02 09:26:19 -07:00
2026-09-12 23:05:29 -07:00
2026-10-04 23:33:24 -07:00
2026-08-23 02:55:02 -07:00
2026-09-07 15:08:56 -07:00
2026-09-30 16:02:16 -07:00
2026-09-07 15:08:56 -07:00
2026-09-07 15:08:56 -07:00
2026-06-05 11:35:01 -07:00
2026-09-26 00:41:01 -07:00
2026-09-26 00:58:48 -07:00
2026-10-05 00:32:42 -07:00
2026-09-12 22:00:50 -07:00
2026-08-24 08:54:06 -07:00
2026-09-26 13:07:22 -07:00
2026-08-24 08:54:06 -07:00
2026-09-12 22:00:50 -07:00
2026-10-01 09:45:22 -07:00
2026-08-01 14:31:14 -07:00
2026-08-24 08:54:06 -07:00
2026-09-12 23:05:29 -07:00
2026-09-12 23:05:29 -07:00
2026-09-12 23:05:29 -07:00
2026-09-07 15:08:56 -07:00
2026-08-01 17:27:14 -07:00
2026-09-07 15:08:56 -07:00
2026-09-15 08:41:44 -07:00
2026-10-01 01:32:57 -07:00
2026-10-01 04:39:39 -07:00
2026-08-18 05:49:10 -07:00
2026-10-04 17:37:32 -07:00
2026-09-07 15:08:56 -07:00
2026-09-12 23:05:29 -07:00
2026-09-12 23:05:29 -07:00
2026-09-12 23:05:29 -07:00
2026-09-12 23:05:29 -07:00
2026-08-24 08:54:06 -07:00
2026-04-20 22:14:00 -07:00
2026-09-25 09:09:48 -07:00
2026-10-01 13:07:18 -07:00
2026-08-18 22:27:15 -07:00
2026-09-18 14:11:58 -07:00
2026-09-12 23:05:29 -07:00
2026-09-30 09:49:30 -07:00
2026-09-14 07:31:23 -07:00
2026-09-12 23:05:29 -07:00
2026-09-15 08:41:44 -07:00
2026-09-07 15:08:56 -07:00
2026-09-07 15:08:56 -07:00
2026-09-01 16:09:42 -07:00
2026-09-24 09:22:50 -07:00
2026-09-26 00:54:23 -07:00
2026-09-07 15:08:56 -07:00
2026-09-26 16:16:32 -07:00
2026-09-25 09:09:48 -07:00
2026-09-21 18:02:56 -07:00
2026-05-12 00:00:31 -07:00
2026-09-07 15:08:56 -07:00
2026-09-07 15:08:56 -07:00
2026-08-12 10:01:46 -07:00
2026-08-24 08:54:06 -07:00
2026-09-27 12:50:05 -07:00
2026-09-07 15:08:56 -07:00
2026-07-17 09:03:05 -07:00
2026-09-07 15:08:56 -07:00
vh
c0d3384c68