[gemini] [HEALTH] Surface local inference throughput and freshness in model_health (#76)

channel_directory.json

@@ -1,5 +1,5 @@
 {
-  "updated_at": "2026-03-27T23:55:36.754689",
+  "updated_at": "2026-03-28T00:57:18.155934",
   "platforms": {
     "discord": [
       {

config.yaml

@@ -1,8 +1,8 @@
 model:
-  default: gpt-5.4
-  provider: openai-codex
+  default: hermes4:14b
+  provider: custom
   context_length: 65536
-  base_url: https://chatgpt.com/backend-api/codex
+  base_url: http://localhost:8081/v1
 toolsets:
 - all
 agent:
@@ -188,7 +188,7 @@ custom_providers:
 - name: Local llama.cpp
   base_url: http://localhost:8081/v1
   api_key: none
-  model: auto
+  model: hermes4:14b
 - name: Google Gemini
   base_url: https://generativelanguage.googleapis.com/v1beta/openai
   api_key_env: GEMINI_API_KEY

tests/test_local_runtime_defaults.py

@@ -0,0 +1,21 @@
+from __future__ import annotations
+
+from pathlib import Path
+
+import yaml
+
+
+def test_config_defaults_to_local_llama_cpp_runtime() -> None:
+    config = yaml.safe_load(Path("config.yaml").read_text())
+
+    assert config["model"]["provider"] == "custom"
+    assert config["model"]["default"] == "hermes4:14b"
+    assert config["model"]["base_url"] == "http://localhost:8081/v1"
+
+    local_provider = next(
+        entry for entry in config["custom_providers"] if entry["name"] == "Local llama.cpp"
+    )
+    assert local_provider["model"] == "hermes4:14b"
+
+    assert config["fallback_model"]["provider"] == "custom"
+    assert config["fallback_model"]["model"] == "gemini-2.5-pro"