Files
esh-pfi-infrastructure/stacks
vh d3721034c1 feat(litellm): Worldtree capability aliases (chat-judge, reranker, scalar-judge)
Stand up the gateway-side capability aliases for the role→capability model
indirection (worldtree-dev's transparent-swap direction; operator: no wt-
prefix, reuse the existing summarizer/classifier/gen alias convention).

- chat-judge  -> selene-1-mini-8b (mode chat)  — WT selene-judgment role.
- reranker    -> qwen3-reranker (mode rerank)  — generic name for the cap.
- scalar-judge -> Skywork-Reward-V2 via a pass_through_endpoint to ana-ml2:8003
  (LiteLLM has no reward/pooling MODE, so it's a passthrough, gateway-key-gated;
  consumers hit /scalar-judge/<route> e.g. /score|/pooling|/classify).

Deliberately NO generic `embedding` alias: embedding vectors are model-specific
(not swap-transparent), so that capability stays `qwen3-embedding` — the model-
specific name is the guardrail against treating it as freely swappable. Verified
all three live (chat-judge 200, reranker present, scalar-judge passthrough 200
returning a Skywork reward). Deployed + gateway health-gated.
2026-06-20 09:53:09 -07:00
..
2026-04-29 13:27:12 -07:00