common/hermes
2
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

feat(aux): use Portal /api/nous/recommended-models for auxiliary models

Browse Source 
Wire the auxiliary client (compaction, vision, session search, web extract)
to the Nous Portal's curated recommended-models endpoint when running on
Nous Portal, with a TTL-cached fetch that mirrors how we pull /models for
pricing.

hermes_cli/models.py
  - fetch_nous_recommended_models(portal_base_url, force_refresh=False)
    10-minute TTL cache, keyed per portal URL (staging vs prod don't
    collide).  Public endpoint, no auth required.  Returns {} on any
    failure so callers always get a dict.
  - get_nous_recommended_aux_model(vision, free_tier=None, ...)
    Tier-aware pick from the payload:
      - Paid tier → paidRecommended{Vision,Compaction}Model, falling back
        to freeRecommended* when the paid field is null (common during
        staged rollouts of new paid models).
      - Free tier → freeRecommended* only, never leaks paid models.
    When free_tier is None, auto-detects via the existing
    check_nous_free_tier() helper (already cached 3 min against
    /api/oauth/account).  Detection errors default to paid so we never
    silently downgrade a paying user.

agent/auxiliary_client.py — _try_nous()
  - Replaces the hardcoded xiaomi/mimo free-tier branch with a single call
    to get_nous_recommended_aux_model(vision=vision).
  - Falls back to _NOUS_MODEL (google/gemini-3-flash-preview) when the
    Portal is unreachable or returns a null recommendation.
  - The Portal is now the source of truth for aux model selection; the
    xiaomi allowlist we used to carry is effectively dead.

Tests (15 new)
  - tests/hermes_cli/test_models.py::TestNousRecommendedModels
    Fetch caching, per-portal keying, network failure, force_refresh;
    paid-prefers-paid, paid-falls-to-free, free-never-leaks-paid,
    auto-detect, detection-error → paid default, null/blank modelName
    handling.
  - tests/agent/test_auxiliary_client.py::TestNousAuxiliaryRefresh
    _try_nous honors Portal recommendation for text + vision, falls
    back to google/gemini-3-flash-preview on None or exception.

Behavior won't visibly change today — both tier recommendations currently
point at google/gemini-3-flash-preview — but the moment the Portal ships
a better paid recommendation, subscribers pick it up within 10 minutes
without a Hermes release.
This commit is contained in:
emozilla
2026-04-21 22:53:45 -04:00
committed by Teknium
parent c22f4a76de
commit 29693f9d8e
4 changed files with 417 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

45
agent/auxiliary_client.py
View File

@@ -182,8 +182,6 @@ auxiliary_is_nous: bool = False
# Default auxiliary models per provider
_OPENROUTER_MODEL = "google/gemini-3-flash-preview"
_NOUS_MODEL = "google/gemini-3-flash-preview"
_NOUS_FREE_TIER_VISION_MODEL = "xiaomi/mimo-v2-omni"
_NOUS_FREE_TIER_AUX_MODEL = "xiaomi/mimo-v2-pro"
_NOUS_DEFAULT_BASE_URL = "https://inference-api.nousresearch.com/v1"
_ANTHROPIC_DEFAULT_BASE_URL = "https://api.anthropic.com"
_AUTH_JSON_PATH = get_hermes_home() / "auth.json"
@@ -927,22 +925,35 @@ def _try_nous(vision: bool = False) -> Tuple[Optional[OpenAI], Optional[str]]:
global auxiliary_is_nous
auxiliary_is_nous = True
logger.debug("Auxiliary client: Nous Portal")
if nous.get("source") == "pool":
model = "gemini-3-flash"
else:
model = _NOUS_MODEL
# Free-tier users can't use paid auxiliary models — use the free
# models instead: mimo-v2-omni for vision, mimo-v2-pro for text tasks.
# Paid accounts keep their tier-appropriate models: gemini-3-flash-preview
# for both text and vision tasks.
# Ask the Portal which model it currently recommends for this task type.
# The /api/nous/recommended-models endpoint is the authoritative source:
# it distinguishes paid vs free tier recommendations, and get_nous_recommended_aux_model
# auto-detects the caller's tier via check_nous_free_tier(). Fall back to
# _NOUS_MODEL (google/gemini-3-flash-preview) when the Portal is unreachable
# or returns a null recommendation for this task type.
model = _NOUS_MODEL
try:
from hermes_cli.models import check_nous_free_tier
if check_nous_free_tier():
model = _NOUS_FREE_TIER_VISION_MODEL if vision else _NOUS_FREE_TIER_AUX_MODEL
logger.debug("Free-tier Nous account — using %s for auxiliary/%s",
model, "vision" if vision else "text")
except Exception:
pass
from hermes_cli.models import get_nous_recommended_aux_model
recommended = get_nous_recommended_aux_model(vision=vision)
if recommended:
model = recommended
logger.debug(
"Auxiliary/%s: using Portal-recommended model %s",
"vision" if vision else "text", model,
)
else:
logger.debug(
"Auxiliary/%s: no Portal recommendation, falling back to %s",
"vision" if vision else "text", model,
)
except Exception as exc:
logger.debug(
"Auxiliary/%s: recommended-models lookup failed (%s); "
"falling back to %s",
"vision" if vision else "text", exc, model,
)
if runtime is not None:
api_key, base_url = runtime
else: