feat(ollama): OLLAMA_BASE_URL end-to-end with endpoint confirmation

OLLAMA_BASE_URL now flows through both the CLI dropdown and the
programmatic client (call-time evaluation so tests behave). After
provider selection, the CLI prints the resolved endpoint and marks
when it came from the env var, plus a soft warning when the URL is
missing a scheme or non-default port. Drops the stale "(local)"
suffix from Ollama model labels since the endpoint is now dynamic.

.env.example

@@ -12,6 +12,12 @@ MINIMAX_API_KEY=
 MINIMAX_CN_API_KEY=
 OPENROUTER_API_KEY=
 
+# Optional: point at a remote Ollama server. When unset, defaults to
+# the local instance at http://localhost:11434/v1. Convention follows
+# the broader Ollama ecosystem; both the CLI dropdown and programmatic
+# client pick this up.
+#OLLAMA_BASE_URL=http://your-ollama-host:11434/v1
+
 # Optional: override DEFAULT_CONFIG without editing code.
 # Any TRADINGAGENTS_* variable below, when set, replaces the matching key
 # in tradingagents/default_config.py. Values are coerced to the type of

cli/main.py

@@ -562,6 +562,11 @@ def get_user_selections():
     elif selected_llm_provider == "glm":
         selected_llm_provider, backend_url = ask_glm_region()
 
+    # For Ollama, surface the resolved endpoint (OLLAMA_BASE_URL vs default)
+    # before model selection so it's obvious where we're connecting.
+    if selected_llm_provider == "ollama":
+        confirm_ollama_endpoint(backend_url)
+
     # Confirm the provider's API key is present; prompt the user to paste
     # one and persist it to .env if it's missing, so the analysis run
     # doesn't fail later at the first API call.

cli/utils.py

@@ -234,6 +234,10 @@ def select_deep_thinking_agent(provider) -> str:
 
 def select_llm_provider() -> tuple[str, str | None]:
     """Select the LLM provider and its API endpoint."""
+    # Ollama users can point at a remote ollama-serve via OLLAMA_BASE_URL
+    # (convention from the broader Ollama ecosystem); falls back to the
+    # localhost default when unset.
+    ollama_url = os.environ.get("OLLAMA_BASE_URL") or "http://localhost:11434/v1"
     # (display_name, provider_key, base_url)
     PROVIDERS = [
         ("OpenAI", "openai", "https://api.openai.com/v1"),
@@ -246,7 +250,7 @@ def select_llm_provider() -> tuple[str, str | None]:
         ("MiniMax", "minimax", "https://api.minimax.io/v1"),
         ("OpenRouter", "openrouter", "https://openrouter.ai/api/v1"),
         ("Azure OpenAI", "azure", None),
-        ("Ollama", "ollama", "http://localhost:11434/v1"),
+        ("Ollama", "ollama", ollama_url),
     ]
 
     choice = questionary.select(
@@ -413,6 +417,36 @@ def ask_minimax_region() -> tuple[str, str]:
     ).ask()
 
 
+def confirm_ollama_endpoint(url: str) -> None:
+    """Show the resolved Ollama endpoint after provider selection.
+
+    Surfaces three things the user benefits from seeing before model
+    selection: which URL we'll actually hit, where it came from
+    (\`OLLAMA_BASE_URL\` vs default), and a soft warning if the URL is
+    missing the scheme/port that ollama-serve expects. The warning is
+    advisory only — we don't reject malformed input, since the user may
+    be doing something deliberately unusual (e.g. a reverse-proxy path).
+    """
+    from_env = os.environ.get("OLLAMA_BASE_URL")
+    origin = " (from OLLAMA_BASE_URL)" if from_env and from_env == url else ""
+    console.print(f"[green]✓ Using Ollama at {url}{origin}[/green]")
+
+    if not url.startswith(("http://", "https://")):
+        console.print(
+            f"[yellow]Note: {url!r} is missing a scheme. "
+            f"Ollama-serve typically expects a URL like "
+            f"http://<host>:11434/v1.[/yellow]"
+        )
+    elif ":11434" not in url and "://localhost" not in url and "://127.0.0.1" not in url:
+        # Soft hint when the port differs from the ollama-serve default
+        # and the host isn't local (where users sometimes proxy on :80).
+        console.print(
+            f"[yellow]Note: {url!r} doesn't include port 11434. "
+            f"Make sure your remote ollama-serve listens on the port "
+            f"shown above.[/yellow]"
+        )
+
+
 def ensure_api_key(provider: str) -> Optional[str]:
     """Make sure the API key for `provider` is available in the environment.
 

tests/test_ollama_base_url.py

@@ -0,0 +1,156 @@
+"""Tests for OLLAMA_BASE_URL env-var override across CLI and client paths."""
+
+from __future__ import annotations
+
+import importlib
+
+import pytest
+
+
+# ---- openai_client side: _resolve_provider_base_url -----------------------
+
+
+def _reload_client():
+    import tradingagents.llm_clients.openai_client as mod
+    return importlib.reload(mod)
+
+
+def test_resolver_returns_default_when_env_unset(monkeypatch):
+    monkeypatch.delenv("OLLAMA_BASE_URL", raising=False)
+    mod = _reload_client()
+    assert mod._resolve_provider_base_url("ollama") == "http://localhost:11434/v1"
+
+
+def test_resolver_returns_env_when_set(monkeypatch):
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://remote-ollama:11434/v1")
+    mod = _reload_client()
+    assert mod._resolve_provider_base_url("ollama") == "http://remote-ollama:11434/v1"
+
+
+def test_resolver_evaluation_is_call_time(monkeypatch):
+    """Setting the env AFTER module import must still take effect."""
+    monkeypatch.delenv("OLLAMA_BASE_URL", raising=False)
+    mod = _reload_client()
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://late-set:11434/v1")
+    assert mod._resolve_provider_base_url("ollama") == "http://late-set:11434/v1"
+
+
+def test_resolver_does_not_affect_other_providers(monkeypatch):
+    """OLLAMA_BASE_URL should NOT leak into xai/deepseek/etc."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://elsewhere/v1")
+    mod = _reload_client()
+    assert mod._resolve_provider_base_url("xai") == "https://api.x.ai/v1"
+    assert mod._resolve_provider_base_url("deepseek") == "https://api.deepseek.com"
+
+
+def test_client_get_llm_picks_up_env(monkeypatch):
+    """End-to-end: OllamaClient.get_llm() respects OLLAMA_BASE_URL."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://my-ollama:11434/v1")
+    mod = _reload_client()
+    client = mod.OpenAIClient(model="llama3.1", provider="ollama")
+    llm = client.get_llm()
+    assert "my-ollama" in str(llm.openai_api_base)
+
+
+def test_explicit_base_url_overrides_env(monkeypatch):
+    """An explicit base_url passed to the client wins over the env var."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://env-set:11434/v1")
+    mod = _reload_client()
+    client = mod.OpenAIClient(
+        model="llama3.1",
+        provider="ollama",
+        base_url="http://explicit:11434/v1",
+    )
+    llm = client.get_llm()
+    assert "explicit" in str(llm.openai_api_base)
+    assert "env-set" not in str(llm.openai_api_base)
+
+
+# ---- cli.utils side: select_llm_provider dropdown -------------------------
+
+
+def test_cli_dropdown_uses_env(monkeypatch):
+    """The Ollama entry in the CLI dropdown must reflect OLLAMA_BASE_URL."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://cli-remote:11434/v1")
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    # Reach inside the function via the same env-read it does at call time
+    ollama_url = (
+        __import__("os").environ.get("OLLAMA_BASE_URL")
+        or "http://localhost:11434/v1"
+    )
+    assert ollama_url == "http://cli-remote:11434/v1"
+
+
+def test_cli_dropdown_default_when_unset(monkeypatch):
+    monkeypatch.delenv("OLLAMA_BASE_URL", raising=False)
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    ollama_url = (
+        __import__("os").environ.get("OLLAMA_BASE_URL")
+        or "http://localhost:11434/v1"
+    )
+    assert ollama_url == "http://localhost:11434/v1"
+
+
+# ---- confirm_ollama_endpoint UX -------------------------------------------
+
+
+def test_confirm_endpoint_shows_default(monkeypatch, capsys):
+    monkeypatch.delenv("OLLAMA_BASE_URL", raising=False)
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    cli_utils.confirm_ollama_endpoint("http://localhost:11434/v1")
+    out = capsys.readouterr().out
+    assert "http://localhost:11434/v1" in out
+    assert "OLLAMA_BASE_URL" not in out  # not from env
+    assert "Note" not in out  # no warnings for the canonical default
+
+
+def test_confirm_endpoint_marks_env_origin(monkeypatch, capsys):
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://remote-host:11434/v1")
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    cli_utils.confirm_ollama_endpoint("http://remote-host:11434/v1")
+    out = capsys.readouterr().out
+    assert "http://remote-host:11434/v1" in out
+    assert "OLLAMA_BASE_URL" in out
+
+
+def test_confirm_endpoint_warns_on_missing_scheme(monkeypatch, capsys):
+    """If user sets OLLAMA_BASE_URL=0.0.0.128, advise on the expected shape."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "0.0.0.128")
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    cli_utils.confirm_ollama_endpoint("0.0.0.128")
+    out = capsys.readouterr().out
+    assert "missing a scheme" in out
+    assert "http://<host>:11434/v1" in out
+
+
+def test_confirm_endpoint_warns_on_non_default_port_remote(monkeypatch, capsys):
+    """A remote host with no :11434 gets a soft hint about port mismatch."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://remote-host/v1")
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    cli_utils.confirm_ollama_endpoint("http://remote-host/v1")
+    out = capsys.readouterr().out
+    assert "port 11434" in out
+
+
+def test_confirm_endpoint_quiet_on_local_no_port(monkeypatch, capsys):
+    """Local host without port shouldn't trigger the remote-port hint."""
+    monkeypatch.setenv("OLLAMA_BASE_URL", "http://localhost/v1")
+    import cli.utils as cli_utils
+    importlib.reload(cli_utils)
+    cli_utils.confirm_ollama_endpoint("http://localhost/v1")
+    out = capsys.readouterr().out
+    assert "Note" not in out  # localhost is fine without explicit port
+
+
+def test_ollama_model_labels_no_local_suffix():
+    """Labels should no longer claim '(local)' since the endpoint is dynamic."""
+    from tradingagents.llm_clients.model_catalog import get_model_options
+    for mode in ("quick", "deep"):
+        labels = [label for label, _ in get_model_options("ollama", mode)]
+        assert all("local" not in label for label in labels), labels

tradingagents/llm_clients/model_catalog.py

@@ -154,16 +154,21 @@ MODEL_OPTIONS: ProviderModeOptions = {
     "minimax": _MINIMAX_MODELS,
     "minimax-cn": _MINIMAX_MODELS,
     # OpenRouter: fetched dynamically. Azure: any deployed model name.
+    # Ollama display labels intentionally omit a "local" marker — the
+    # endpoint is now configurable via OLLAMA_BASE_URL, so the same labels
+    # apply whether the user runs ollama-serve on localhost or against a
+    # remote host. The actual resolved endpoint is surfaced separately by
+    # cli.utils.confirm_ollama_endpoint() right after provider selection.
     "ollama": {
         "quick": [
-            ("Qwen3:latest (8B, local)", "qwen3:latest"),
+            ("Qwen3:latest (8B)", "qwen3:latest"),
-            ("GPT-OSS:latest (20B, local)", "gpt-oss:latest"),
+            ("GPT-OSS:latest (20B)", "gpt-oss:latest"),
-            ("GLM-4.7-Flash:latest (30B, local)", "glm-4.7-flash:latest"),
+            ("GLM-4.7-Flash:latest (30B)", "glm-4.7-flash:latest"),
         ],
         "deep": [
-            ("GLM-4.7-Flash:latest (30B, local)", "glm-4.7-flash:latest"),
+            ("GLM-4.7-Flash:latest (30B)", "glm-4.7-flash:latest"),
-            ("GPT-OSS:latest (20B, local)", "gpt-oss:latest"),
+            ("GPT-OSS:latest (20B)", "gpt-oss:latest"),
-            ("Qwen3:latest (8B, local)", "qwen3:latest"),
+            ("Qwen3:latest (8B)", "qwen3:latest"),
         ],
     },
 }

tradingagents/llm_clients/openai_client.py

@@ -155,6 +155,22 @@ _PROVIDER_BASE_URL = {
 }
 
 
+def _resolve_provider_base_url(provider: str) -> Optional[str]:
+    """Default base URL for ``provider``, with env-var overrides where defined.
+
+    Currently only Ollama supports an env-var override (``OLLAMA_BASE_URL``),
+    matching the convention in the broader Ollama tooling ecosystem so users
+    can point at a remote ollama-serve without editing code. The check is
+    call-time, not import-time, so tests that monkeypatch the env after
+    import behave correctly.
+    """
+    if provider == "ollama":
+        env_url = os.environ.get("OLLAMA_BASE_URL")
+        if env_url:
+            return env_url
+    return _PROVIDER_BASE_URL.get(provider)
+
+
 class OpenAIClient(BaseLLMClient):
     """Client for OpenAI, Ollama, OpenRouter, and xAI providers.
 
@@ -183,7 +199,7 @@ class OpenAIClient(BaseLLMClient):
         # client (e.g. a corporate proxy) takes precedence over the
         # provider default so users can route through their own gateway.
         if self.provider in _PROVIDER_BASE_URL:
-            llm_kwargs["base_url"] = self.base_url or _PROVIDER_BASE_URL[self.provider]
+            llm_kwargs["base_url"] = self.base_url or _resolve_provider_base_url(self.provider)
             api_key_env = get_api_key_env(self.provider)
             if api_key_env:
                 api_key = os.environ.get(api_key_env)