feat(setup,chat): detect actually-loaded local model via provider-specific API

For Ollama, /api/tags returns all installed models, not running ones.
Add fetch_loaded_models() using /api/ps for Ollama (and /v1/models for
LM Studio/llama.cpp, which already return only loaded models).

_show_local_model_status() now calls fetch_loaded_models() so the
setup wizard correctly shows only in-memory models for Ollama.

At chat session startup, local providers warn when the configured model
is not currently loaded, or when nothing is loaded at all.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/pyra/chat/session.py

@@ -25,6 +25,7 @@ from pyra.plugins.executor import ToolExecutor
 from pyra.plugins.registry import PluginRegistry
 from pyra.security.injection import scan_response
 from pyra.setup.providers import get_provider
+from pyra.setup.wizard import fetch_loaded_models
 from pyra.utils.paths import pyra_home
 
 _HISTORY_FILE = pyra_home() / ".chat_history"
@@ -176,6 +177,16 @@ def start_chat() -> None:
         "[dim]Type /help for commands, /quit to exit.[/dim]"
     )
 
+    if provider.group == "Local":
+        loaded = fetch_loaded_models(provider)
+        if not loaded:
+            render_info(f"No model currently loaded in {provider.display_name}.")
+        elif cfg.ai.model not in loaded:
+            render_info(
+                f"Model '{cfg.ai.model}' not loaded in {provider.display_name}. "
+                f"Loaded: {', '.join(loaded)}"
+            )
+
     _flags: dict = {"use_tools": True}
 
     while True:

src/pyra/setup/wizard.py

@@ -377,9 +377,26 @@ def _check_local_server(provider: Provider) -> None:
             # "retry" → loop
 
 
+def fetch_loaded_models(provider: Provider) -> list[str]:
+    """Return models currently loaded in RAM from a local provider's API."""
+    if not provider.base_url:
+        return []
+    try:
+        if provider.id == "ollama":
+            resp = httpx.get(f"{provider.base_url}/api/ps", timeout=3.0)
+            resp.raise_for_status()
+            return [m["name"] for m in resp.json().get("models", [])]
+        else:
+            resp = httpx.get(f"{provider.base_url}/models", timeout=3.0)
+            resp.raise_for_status()
+            return [m["id"] for m in resp.json().get("data", [])]
+    except Exception:
+        return []
+
+
 def _show_local_model_status(provider: Provider) -> None:
     """Print a one-line status showing which models are currently loaded."""
-    models = _fetch_local_models(provider)
+    models = fetch_loaded_models(provider)
     if not models:
         console.print("  [yellow]No model currently loaded[/yellow]")
     elif len(models) == 1:

tests/unit/test_setup_wizard.py

@@ -377,12 +377,57 @@ def test_draft_file_has_correct_permissions(tmp_pyra_home):
         assert mode == "600"
 
 
+# ── fetch_loaded_models ───────────────────────────────────────────────────────
+
+def test_fetch_loaded_models_ollama_uses_api_ps(monkeypatch):
+    import pyra.setup.wizard as wiz
+    from pyra.setup.providers import get_provider
+    mock_resp = MagicMock()
+    mock_resp.json.return_value = {"models": [{"name": "llama3:latest"}, {"name": "mistral"}]}
+    mock_resp.raise_for_status = lambda: None
+    calls = []
+    monkeypatch.setattr(wiz.httpx, "get", lambda url, **kw: (calls.append(url), mock_resp)[1])
+    result = wiz.fetch_loaded_models(get_provider("ollama"))
+    assert result == ["llama3:latest", "mistral"]
+    assert any("/api/ps" in u for u in calls)
+
+
+def test_fetch_loaded_models_lmstudio_uses_models_endpoint(monkeypatch):
+    import pyra.setup.wizard as wiz
+    from pyra.setup.providers import get_provider
+    mock_resp = MagicMock()
+    mock_resp.json.return_value = {"data": [{"id": "gemma-4b"}]}
+    mock_resp.raise_for_status = lambda: None
+    calls = []
+    monkeypatch.setattr(wiz.httpx, "get", lambda url, **kw: (calls.append(url), mock_resp)[1])
+    result = wiz.fetch_loaded_models(get_provider("lmstudio"))
+    assert result == ["gemma-4b"]
+    assert any("/models" in u for u in calls)
+
+
+def test_fetch_loaded_models_returns_empty_on_error(monkeypatch):
+    import pyra.setup.wizard as wiz
+    from pyra.setup.providers import get_provider
+    monkeypatch.setattr(wiz.httpx, "get", MagicMock(side_effect=Exception("conn refused")))
+    assert wiz.fetch_loaded_models(get_provider("ollama")) == []
+
+
+def test_fetch_loaded_models_returns_empty_when_no_base_url():
+    import pyra.setup.wizard as wiz
+    from pyra.setup.providers import Provider
+    provider = Provider(
+        id="test", display_name="Test", requires_key=False,
+        default_model="x", litellm_prefix="openai/", group="Local",
+    )
+    assert wiz.fetch_loaded_models(provider) == []
+
+
 # ── _show_local_model_status ──────────────────────────────────────────────────
 
 def test_show_local_model_status_one_model(monkeypatch):
     import pyra.setup.wizard as wiz
     from pyra.setup.providers import get_provider
-    monkeypatch.setattr(wiz, "_fetch_local_models", lambda p: ["gemma-4b"])
+    monkeypatch.setattr(wiz, "fetch_loaded_models", lambda p: ["gemma-4b"])
     printed = []
     monkeypatch.setattr(wiz.console, "print", lambda *a, **kw: printed.append(str(a)))
     wiz._show_local_model_status(get_provider("lmstudio"))
@@ -392,7 +437,7 @@ def test_show_local_model_status_one_model(monkeypatch):
 def test_show_local_model_status_none(monkeypatch):
     import pyra.setup.wizard as wiz
     from pyra.setup.providers import get_provider
-    monkeypatch.setattr(wiz, "_fetch_local_models", lambda p: [])
+    monkeypatch.setattr(wiz, "fetch_loaded_models", lambda p: [])
     printed = []
     monkeypatch.setattr(wiz.console, "print", lambda *a, **kw: printed.append(str(a)))
     wiz._show_local_model_status(get_provider("lmstudio"))
@@ -402,7 +447,7 @@ def test_show_local_model_status_none(monkeypatch):
 def test_show_local_model_status_multiple(monkeypatch):
     import pyra.setup.wizard as wiz
     from pyra.setup.providers import get_provider
-    monkeypatch.setattr(wiz, "_fetch_local_models", lambda p: ["a", "b", "c"])
+    monkeypatch.setattr(wiz, "fetch_loaded_models", lambda p: ["a", "b", "c"])
     printed = []
     monkeypatch.setattr(wiz.console, "print", lambda *a, **kw: printed.append(str(a)))
     wiz._show_local_model_status(get_provider("lmstudio"))