fix(anthropic): resolve the count-tokens base like chat when the deployment has none

litellm/llms/anthropic/count_tokens/transformation.py

@@ -7,7 +7,7 @@ This module handles the transformation of requests to Anthropic's CountTokens AP
 from typing import Any, Final
 
 from litellm.constants import ANTHROPIC_TOKEN_COUNTING_BETA_VERSION
-from litellm.llms.anthropic.wif import anthropic_base_without_chat_suffix
+from litellm.llms.anthropic.wif import resolve_anthropic_base
 
 
 class AnthropicCountTokensConfig:
@@ -27,14 +27,14 @@ class AnthropicCountTokensConfig:
         Args:
             api_base: The deployment's api_base, which names the chat surface (a host, or a
                 base already carrying ``/v1`` or ``/v1/messages``); the count-tokens path is
-                appended to it, so it is never the full count-tokens URL
+                appended to it, so it is never the full count-tokens URL. Unset or empty falls
+                back to ``ANTHROPIC_API_BASE`` / ``ANTHROPIC_BASE_URL`` and then Anthropic's
+                host, the same resolution chat and the federated exchange use
 
         Returns:
             The endpoint URL for the CountTokens API
         """
-        if api_base is None:
-            return "https://api.anthropic.com/v1/messages/count_tokens"
-        return anthropic_base_without_chat_suffix(api_base) + "/v1/messages/count_tokens"
+        return resolve_anthropic_base(api_base) + "/v1/messages/count_tokens"
 
     def transform_request_to_count_tokens(
         self,

litellm/llms/anthropic/wif.py

@@ -254,7 +254,7 @@ def get_anthropic_wif_token(
     params: Final = resolve_anthropic_wif_params(litellm_params)
     if params is None:
         return None
-    exchange_base: Final = _token_exchange_base(api_base)
+    exchange_base: Final = resolve_anthropic_base(api_base)
     _raise_if_exchange_host_untrusted(exchange_base, model)
     result: Final = engine.get_token(build_anthropic_wif_spec(params, exchange_base))
     return _token_from_result(result, model, params)
@@ -269,7 +269,7 @@ async def aget_anthropic_wif_token(
     params: Final = resolve_anthropic_wif_params(litellm_params)
     if params is None:
         return None
-    exchange_base: Final = _token_exchange_base(api_base)
+    exchange_base: Final = resolve_anthropic_base(api_base)
     _raise_if_exchange_host_untrusted(exchange_base, model)
     result: Final = await engine.aget_token(build_anthropic_wif_spec(params, exchange_base))
     return _token_from_result(result, model, params)
@@ -283,11 +283,12 @@ def _token_from_result(result: ExchangeResult, model: str, params: AnthropicWifP
             _raise_anthropic_wif_error(result, model=model, workspace_id_set=params.workspace_id is not None)
 
 
-def _token_exchange_base(api_base: str | None) -> str:
-    """Exchange base for any caller-supplied form of the deployment base: trailing
-    slashes and chat-appended ``/v1/messages`` suffixes stripped, so every tier
-    derives the same token URL (and cache key) for the same deployment."""
-    return anthropic_base_without_chat_suffix(api_base if api_base is not None else _resolve_default_api_base())
+def resolve_anthropic_base(api_base: str | None) -> str:
+    """The base every Anthropic tier derives its URLs from: the deployment api_base when set,
+    else ``ANTHROPIC_API_BASE`` / ``ANTHROPIC_BASE_URL``, else Anthropic's host, with trailing
+    slashes and chat-appended ``/v1/messages`` suffixes stripped, so the token URL, the cache key
+    and the count-tokens URL all agree for the same deployment."""
+    return anthropic_base_without_chat_suffix(api_base or _resolve_default_api_base())
 
 
 def _trusted_exchange_hosts() -> frozenset[str]:

tests/test_litellm/llms/anthropic/test_anthropic_count_tokens_transformation.py

@@ -99,16 +99,41 @@ def test_transform_no_system_no_tools():
     ("api_base", "expected"),
     [
         (None, "https://api.anthropic.com/v1/messages/count_tokens"),
+        ("", "https://api.anthropic.com/v1/messages/count_tokens"),
         ("https://gateway.example", "https://gateway.example/v1/messages/count_tokens"),
         ("https://gateway.example/", "https://gateway.example/v1/messages/count_tokens"),
         ("https://gateway.example/v1", "https://gateway.example/v1/messages/count_tokens"),
         ("https://gateway.example/anthropic/v1/messages", "https://gateway.example/anthropic/v1/messages/count_tokens"),
     ],
 )
-def test_endpoint_appends_count_tokens_path_to_deployment_api_base(api_base, expected):
+def test_endpoint_appends_count_tokens_path_to_deployment_api_base(api_base, expected, monkeypatch):
+    monkeypatch.delenv("ANTHROPIC_API_BASE", raising=False)
+    monkeypatch.delenv("ANTHROPIC_BASE_URL", raising=False)
     assert AnthropicCountTokensConfig().get_anthropic_count_tokens_endpoint(api_base) == expected
 
 
+@pytest.mark.parametrize("env_name", ["ANTHROPIC_API_BASE", "ANTHROPIC_BASE_URL"])
+@pytest.mark.parametrize("api_base", [None, ""])
+def test_endpoint_without_deployment_api_base_follows_env_base(env_name, api_base, monkeypatch):
+    """Chat and the federated exchange resolve an unset deployment base through the environment,
+    so an env-only gateway must receive the count too, never Anthropic's public host."""
+    monkeypatch.delenv("ANTHROPIC_API_BASE", raising=False)
+    monkeypatch.delenv("ANTHROPIC_BASE_URL", raising=False)
+    monkeypatch.setenv(env_name, "https://env-gateway.example/v1/messages/")
+    assert (
+        AnthropicCountTokensConfig().get_anthropic_count_tokens_endpoint(api_base)
+        == "https://env-gateway.example/v1/messages/count_tokens"
+    )
+
+
+def test_endpoint_prefers_deployment_api_base_over_env_base(monkeypatch):
+    monkeypatch.setenv("ANTHROPIC_API_BASE", "https://env-gateway.example")
+    assert (
+        AnthropicCountTokensConfig().get_anthropic_count_tokens_endpoint("https://gateway.example/v1")
+        == "https://gateway.example/v1/messages/count_tokens"
+    )
+
+
 @pytest.fixture
 def httpx_transport_clients(monkeypatch):
     monkeypatch.setattr(litellm, "disable_aiohttp_transport", True)

tests/test_litellm/llms/anthropic/test_anthropic_wif.py

@@ -303,6 +303,11 @@ class TestBaseUrlDerivation:
         monkeypatch.setenv("ANTHROPIC_API_BASE", "https://env.example.com")
         assert self._mint(None, monkeypatch) == "https://env.example.com/v1/oauth/token"
 
+    def test_empty_api_base_falls_back_like_unset(self, monkeypatch: pytest.MonkeyPatch):
+        """Chat treats an empty deployment api_base as unset; the exchange must not refuse host ''."""
+        monkeypatch.setenv("ANTHROPIC_API_BASE", "https://env.example.com")
+        assert self._mint("", monkeypatch) == "https://env.example.com/v1/oauth/token"
+
     def test_env_base_url(self, monkeypatch: pytest.MonkeyPatch):
         monkeypatch.setenv("ANTHROPIC_BASE_URL", "https://base.example.com")
         assert self._mint(None, monkeypatch) == "https://base.example.com/v1/oauth/token"