feat(api-server): per-client model routing via model_routes (#3176 salvage)

Adds a no-code routing layer to the OpenAI-compatible API server so one
Hermes deployment can map different API clients to different
model/provider backends. Clients pick a backend by sending a configured
alias as the OpenAI 'model' field; unmatched values fall back to the
global model. Configured aliases are listed by GET /v1/models.

Precedence (highest first): session /model override > model_routes
route > global config. Route provider credentials resolve through
_resolve_runtime_agent_kwargs_for_provider (same seam as
channel_overrides); per-route api_key/base_url are upstream provider
credential overrides — never caller auth, never logged.

Salvaged and rebased from PR #3176 by @Mibayy onto current main.
This commit is contained in:
Mibayy 2026-07-02 04:47:12 -07:00 committed by Teknium
parent ce9aa869fc
commit 4a09b692ec
4 changed files with 503 additions and 16 deletions

View file

@ -46,6 +46,7 @@ ACP_REGISTRY_MANIFEST = REPO_ROOT / "acp_registry" / "agent.json"
# Auto-extracted from noreply emails + manual overrides
AUTHOR_MAP = {
"louis@letsfive.io": "Mibayy", # PR #3243 salvage (/compact alias + preview/aggressive flags for /compress)
"louis@letsfive.io": "Mibayy", # PR #3176 salvage (api-server: per-client model routing via model_routes)
"ai-lab@foxmail.com": "CrazyBoyM", # PR #55828 salvage (image_gen openai-codex: wire image-to-image / reference-image editing via Codex Responses input_image parts; magic-byte + read-guard + 25MB-cap + clamp-to-16 hardening)
"r0gersm1th@users.noreply.github.com": "r0gersm1th", # PR #3219 salvage (whatsapp bridge: resolve LID sender IDs to phone numbers in the message payload so phone-based allowlists match; commit authored by collaborator r0gersm1th, PR by @ajmeese7)
"louis@letsfive.io": "Mibayy", # PR #3296 salvage (status: provider label honors config.yaml model.base_url, not just OPENAI_BASE_URL env)