rblundon/homelab
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

hindsight: re-promote upstage/solar-pro4:free as primary LLM (ingress timeout fixed)

Browse Source 
Ingress proxy timeout now 300s (commit 152230c), so a slow agentic reflect
(~132s) completes before nginx cuts the connection. solar-pro4:free passed
tool-calling + grounded-output probes (t_d0dffc3d). BASE_URL/PROVIDER
unchanged; fallback comment -> stepfun/step-3.7-flash:free.

Refs t_e3375410.
This commit is contained in:
Hermes Agent service account
2026-08-25 12:05:54 -05:00
parent 152230c100
commit 9bfc9384e4
1 changed files with 6 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
cluster/applications/hindsight/values.yaml
View File

@@ -14,8 +14,8 @@
# HINDSIGHT_API_DATABASE_URL (verified with a live envFrom test pod). # HINDSIGHT_API_DATABASE_URL (verified with a live envFrom test pod).
# - LLM is the Nous free-tier inference endpoint # - LLM is the Nous free-tier inference endpoint
# (https://inference-api.nousresearch.com/v1), model # (https://inference-api.nousresearch.com/v1), model
# `stepfun/step-3.7-flash:free` (mandatory reasoning, 262K ctx). Fallback # `upstage/solar-pro4:free` (tool-calling, verified reflect). Fallback
# (documented, NOT deployed): `solar-pro4:free`. API key via existingSecret # (documented, NOT deployed): `stepfun/step-3.7-flash:free`. API key via existingSecret
# envFrom (hindsight-credentials / HINDSIGHT_API_LLM_API_KEY), sourced from # envFrom (hindsight-credentials / HINDSIGHT_API_LLM_API_KEY), sourced from
# 1Password `nous` item per decision 4. # 1Password `nous` item per decision 4.
# - Ingress is driven through the chart's NATIVE ingress template (approved # - Ingress is driven through the chart's NATIVE ingress template (approved
@@ -57,8 +57,10 @@ api:
env: env:
HINDSIGHT_API_LLM_BASE_URL: "https://inference-api.nousresearch.com/v1" HINDSIGHT_API_LLM_BASE_URL: "https://inference-api.nousresearch.com/v1"
HINDSIGHT_API_LLM_PROVIDER: "openai" HINDSIGHT_API_LLM_PROVIDER: "openai"
# Primary LLM. Fallback (documented, NOT deployed): solar-pro4:free # Primary LLM. Timeout raised to 300s (ingress) so a slow agentic reflect
HINDSIGHT_API_LLM_MODEL: "stepfun/step-3.7-flash:free" # (~132s on solar-pro4) completes; verified tool-calling + grounded output.
HINDSIGHT_API_LLM_MODEL: "upstage/solar-pro4:free"
# Fallback (documented, NOT deployed): stepfun/step-3.7-flash:free
# ---------------------------------------------------------------------------- # ----------------------------------------------------------------------------
# Ingress via the chart's native template. # Ingress via the chart's native template.