scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-12 23:01:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 89
litellm/ui
History
Exact
Tin Chi Lo 6fe3cc2521 feat(router): add best_of_n router with parallel fan-out and synthesizer 
A best_of_n/<name> deployment fans each request out to N configured model
groups in parallel, then a configured synthesizer merges the candidates into
the final answer (or picks the best candidate verbatim when tools are in
play). Registered as a custom provider so chat completions, /v1/messages and
/v1/responses all reach it through the existing bridges, streaming included.
Arms and the synthesizer each log their own spend under the caller's key with
best_of_n internal-call origins; the parent request logs an explicit zero.
Requested in #35781 and #30456
2026-09-01 15:07:07 -07:00
..
litellm-dashboard feat(router): add best_of_n router with parallel fan-out and synthesizer 2026-09-01 15:07:07 -07:00
Dockerfile build(ui): bump nginx to 1.31-alpine 2026-08-27 17:09:17 +00:00
nginx.conf fix(ui): boot the UI image as an arbitrary uid by anchoring nginx writes under /tmp (#37982) 2026-08-24 11:57:36 -07:00