3.4 KiB
title, type, status, created, updated, verified_on, confidence, tags, sources
| title | type | status | created | updated | verified_on | confidence | tags | sources | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model Routing | concept | active | 2026-07-22 | 2026-10-20 | 2026-10-20 | high |
|
|
Model Routing
Purpose
How Hermes selects providers, applies combos, and enforces context limits.
Current Flow
- Main path:
custom:litellm→http://litellm:4000/v1 - Default model:
freellmapi-auto - OmniRoute used for the default chat path (
custom:omni-101/smart-route, since 2026-10-24 19:06) and for all eight auxiliary slots (approval,compression,curator,mcp,profile_describer,skills_hub,title_generation,vision— allno-think/freeapi/autoas of 2026-10-25 19:13)
Live
model:block re-read 2026-10-26 — the default has moved TWICE.model.defaultissmart-routeon providercustom:omni-101(http://100.78.63.20:20128),api_mode: chat_completions. It moved tonous/meituan/longcat-2.5-preview:freeon 2026-10-20 22:01 and back on 2026-10-24 19:06:10. A six-provider top-levelfallback_providerschain is live again, headed by the deadnouscredential (Invalid refresh tokensince 2026-10-22). → systems/current-profile · incidents/2026-10-22-nous-portal-refresh-token-invalid
The five
custom_providersand their live base URLs:
Name Base URL Host litellmhttp://litellm:4000/v1LiteLLM proxy (m93p CT500) guacanohttp://ollama:11000/v1Guanaco on CT205 Omniroute:20128http://omniroute:20128/v1OmniRoute on m93p CT101 command-codehttps://api.commandcode.ai/provider/v1remote omni-101http://100.78.63.20:20128OmniRoute node omniroute-1(direct IP, not a DNS name)Note the two distinct OmniRoute endpoints:
Omniroute:20128(DNS, m93p CT101) andomni-101(raw IP100.78.63.20:20128).omni-101is the pin on effectively every model-calling cron job, and it has been the single point of failure in the 10-14, 10-16/17 and 10-19 incidents — its/api/healthreads200 {"status":"ok"}throughout, so reachability probes do not detect that failure class; only job-level error text does.
Context Enforcement
- Three layers: model native → combo
data.config.context_length→ Hermes client cap - Effective context is the minimum of the three
Combo Behavior
priorityretries same provider up tomaxRetries, then errors; it does NOT fall throughfill-firstwalks the list on failureauto/lkgprecalculates per request using health scoring- Same-provider entries in a combo share rate-limit buckets
- Session affinity caches provider in
session_model_history
Aliases and Mappings
- Model aliases resolved via
key_valuetable (modelAliasesnamespace) - Wildcard mappings via
model_combo_mappings - OpenCode/Zen entries may resolve back to
opencode-zenprovider internally
History
- User-provided: OpenRouter was used previously; provider lists now centered on LiteLLM + OmniRoute
- User-provided:
persist_switch_by_default=false;/modelis per-session unless--globalused - Inferred: Configuration undergoes periodic corrections related to vision, model fallback, and browser backend tuning