test(e2e): source the cheap Anthropic model from e2e_config across suites

tests/e2e/llm_translation/test_chat_completions_regression_e2e.py

@@ -13,7 +13,7 @@ from __future__ import annotations
 
 import pytest
 
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import unwrap
 from models import ChatBody, ChatMessage
 from passthrough_client import PassthroughClient
@@ -22,7 +22,7 @@ pytestmark = pytest.mark.e2e
 
 CHAT_MODELS: tuple[tuple[str, str], ...] = (
     ("gpt-5.5", "openai"),
-    ("claude-haiku-4-5", "anthropic"),
+    (CHEAP_ANTHROPIC_MODEL, "anthropic"),
     ("gemini-2.5-flash", "gemini"),
 )
 

tests/e2e/llm_translation/test_passthrough_e2e.py

@@ -13,7 +13,7 @@ A passthrough call returning non-2xx fails hard (never a skip); once it returns
 
 import pytest
 
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import StreamingResponse, require_successful_call
 from models import SpendLogRow
 from passthrough_client import (
@@ -112,7 +112,7 @@ def test_gemini_passthrough_tool_call_logs_cost(
 def test_anthropic_passthrough_nonstreaming_logs_cost(
     client: PassthroughClient, scoped_key: str
 ) -> None:
-    result = client.anthropic_message(scoped_key, "claude-haiku-4-5", "Say hello")
+    result = client.anthropic_message(scoped_key, CHEAP_ANTHROPIC_MODEL, "Say hello")
     require_successful_call(result)
 
     row = _fetch_cost_breakdown(client, result)
@@ -124,7 +124,7 @@ def test_anthropic_passthrough_streaming_logs_cost(
     client: PassthroughClient, scoped_key: str
 ) -> None:
     result = client.anthropic_message(
-        scoped_key, "claude-haiku-4-5", "Count to five", stream=True
+        scoped_key, CHEAP_ANTHROPIC_MODEL, "Count to five", stream=True
     )
     require_successful_call(result)
     assert result.chunks > 0, "streaming passthrough produced no events"
@@ -138,7 +138,7 @@ def test_anthropic_passthrough_tool_call_logs_cost(
 ) -> None:
     result = client.anthropic_message(
         scoped_key,
-        "claude-haiku-4-5",
+        CHEAP_ANTHROPIC_MODEL,
         "What is the weather in Paris? Use the get_weather tool.",
         tools=[
             AnthropicTool(

tests/e2e/quota_management/budgets/test_budget_enforcement_e2e.py

@@ -17,7 +17,7 @@ from typing import Callable, List, Type
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import run_case
 
@@ -31,7 +31,7 @@ def _assert_budget_blocks(client: BudgetClient, key: str, *, user: str = "") ->
     for _ in range(40):
         result = client.chat(
             key,
-            "claude-haiku-4-5",
+            CHEAP_ANTHROPIC_MODEL,
             f"spend {unique_marker()}",
             max_tokens=16,
             user=user or None,
@@ -88,7 +88,7 @@ class EndUserBudgetCase(_BudgetCase):
         customer = f"e2e-budget-cust-{unique_marker()}"
         self.client.create_customer(customer, max_budget=3e-6)
         self._undo.append(lambda: self.client.delete_customers([customer]))
-        self.key = self.client.generate_key(models=["claude-haiku-4-5"])
+        self.key = self.client.generate_key(models=[CHEAP_ANTHROPIC_MODEL])
         self._undo.append(lambda: self.client.delete_key(self.key))
         self._customer = customer
 

tests/e2e/quota_management/budgets/test_budget_fallback_e2e.py

@@ -9,13 +9,13 @@ import time
 import pytest
 
 from budget_client import BudgetClient, model_budget
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from lifecycle import ResourceManager
 from models import AnthropicMessagesResponse
 
 pytestmark = pytest.mark.e2e
 
-PRIMARY_MODEL = "claude-haiku-4-5"
+PRIMARY_MODEL = CHEAP_ANTHROPIC_MODEL
 FALLBACK_MODEL = "gpt-5.5"
 
 

tests/e2e/quota_management/budgets/test_budget_reset_advances_e2e.py

@@ -19,7 +19,7 @@ from datetime import datetime
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 from models import BudgetWindow
@@ -32,7 +32,7 @@ TINY_CAP = 3e-6
 
 
 def _call(client: BudgetClient, key: str):
-    return client.chat(key, "claude-haiku-4-5", f"advance {unique_marker()}", max_tokens=16)
+    return client.chat(key, CHEAP_ANTHROPIC_MODEL, f"advance {unique_marker()}", max_tokens=16)
 
 
 def _as_datetime(value: str) -> datetime:

tests/e2e/quota_management/budgets/test_budget_reset_e2e.py

@@ -13,7 +13,7 @@ import time
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 
@@ -22,7 +22,7 @@ pytestmark = pytest.mark.e2e
 
 def _call(client: BudgetClient, key: str):
     return client.chat(
-        key, "claude-haiku-4-5", f"reset {unique_marker()}", max_tokens=16
+        key, CHEAP_ANTHROPIC_MODEL, f"reset {unique_marker()}", max_tokens=16
     )
 
 

tests/e2e/quota_management/budgets/test_model_max_budget_e2e.py

@@ -11,13 +11,13 @@ import time
 import pytest
 
 from budget_client import BudgetClient, is_budget_block, model_budget
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 
 pytestmark = pytest.mark.e2e
 
-CAPPED_MODEL = "claude-haiku-4-5"
+CAPPED_MODEL = CHEAP_ANTHROPIC_MODEL
 FREE_MODEL = "gemini-2.5-flash"
 
 

tests/e2e/quota_management/budgets/test_multi_window_budget_e2e.py

@@ -13,7 +13,7 @@ import time
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 from models import BudgetWindow
@@ -25,7 +25,7 @@ WINDOW_SECONDS = 30  # the tight window; calls succeed again only after it elaps
 
 def _call(client: BudgetClient, key: str):
     return client.chat(
-        key, "claude-haiku-4-5", f"window {unique_marker()}", max_tokens=16
+        key, CHEAP_ANTHROPIC_MODEL, f"window {unique_marker()}", max_tokens=16
     )
 
 

tests/e2e/quota_management/budgets/test_soft_budget_e2e.py

@@ -10,7 +10,7 @@ load-bearing behavior: soft != block.
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 
@@ -27,7 +27,7 @@ def test_soft_budget_does_not_block(
 
     for _ in range(3):
         result = client.chat(
-            key, "claude-haiku-4-5", f"hi {unique_marker()}", max_tokens=16
+            key, CHEAP_ANTHROPIC_MODEL, f"hi {unique_marker()}", max_tokens=16
         )
         assert not is_budget_block(result), (
             "soft_budget blocked a request; it must alert only, not block "

tests/e2e/quota_management/budgets/test_spend_counter_reseed_e2e.py

@@ -26,7 +26,7 @@ import pytest
 from pydantic import TypeAdapter, ValidationError
 
 from budget_client import BudgetClient
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import StreamingResponse
 from lifecycle import ResourceManager
 
@@ -36,7 +36,7 @@ if TYPE_CHECKING:
 
 pytestmark = pytest.mark.e2e
 
-MODEL = "claude-haiku-4-5"
+MODEL = CHEAP_ANTHROPIC_MODEL
 ACCUMULATE_CALLS = 24
 BURST = 6
 # proxy_batch_write_at (60s) flushes the spend to the DB and default_redis_ttl (20s)

tests/e2e/quota_management/budgets/test_tag_budget_e2e.py

@@ -11,7 +11,7 @@ import time
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 
@@ -23,7 +23,7 @@ TINY_BUDGET = 1e-6
 def _tagged_call(client: BudgetClient, key: str, tag: str):
     result = client.chat(
         key,
-        "claude-haiku-4-5",
+        CHEAP_ANTHROPIC_MODEL,
         f"hi {unique_marker()}",
         tags=[tag],
         max_tokens=64,

tests/e2e/quota_management/budgets/test_team_member_budget_e2e.py

@@ -19,14 +19,14 @@ from dataclasses import dataclass
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import Success, require_successful_call
 from lifecycle import ResourceManager
 from models import ChatBody, ChatMessage
 
 pytestmark = pytest.mark.e2e
 
-MODEL = "claude-haiku-4-5"
+MODEL = CHEAP_ANTHROPIC_MODEL
 TEAM_BUDGET = 100.0
 MEMBER_BUDGET = 3e-6
 BURST = 6

tests/e2e/quota_management/budgets/test_team_member_budget_reset_e2e.py

@@ -4,7 +4,7 @@ from datetime import datetime
 import pytest
 
 from budget_client import BudgetClient
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 
@@ -34,7 +34,7 @@ def test_team_member_budget_reset_keeps_advancing(client: BudgetClient, resource
     # the member can spend within the team while the window is live
     key = client.generate_key(team_id=team_id, user_id=user_id)
     resources.defer(lambda: client.delete_key(key))
-    require_successful_call(client.chat(key, "claude-haiku-4-5", f"reset {unique_marker()}", max_tokens=16))
+    require_successful_call(client.chat(key, CHEAP_ANTHROPIC_MODEL, f"reset {unique_marker()}", max_tokens=16))
 
     # once the window elapses the reset job must move budget_reset_at forward; a job
     # that skips the member's budget row (the #25109 regression) leaves it pinned at

tests/e2e/quota_management/budgets/test_team_multi_window_budget_e2e.py

@@ -18,7 +18,7 @@ import time
 import pytest
 
 from budget_client import BudgetClient, is_budget_block
-from e2e_config import unique_marker
+from e2e_config import CHEAP_ANTHROPIC_MODEL, unique_marker
 from e2e_http import require_successful_call
 from lifecycle import ResourceManager
 from models import BudgetWindow
@@ -29,7 +29,7 @@ WINDOW_SECONDS = 30
 
 
 def _call(client: BudgetClient, key: str):
-    return client.chat(key, "claude-haiku-4-5", f"team-window {unique_marker()}", max_tokens=16)
+    return client.chat(key, CHEAP_ANTHROPIC_MODEL, f"team-window {unique_marker()}", max_tokens=16)
 
 
 @pytest.mark.covers("quota_management.budget.team_multi_window.blocks_then_resets")
@@ -42,7 +42,7 @@ def test_team_short_window_blocks_then_resets(client: BudgetClient, resources: R
         ],
     )
     resources.defer(lambda: client.delete_team(team_id))
-    key = client.generate_key(team_id=team_id, models=["claude-haiku-4-5"])
+    key = client.generate_key(team_id=team_id, models=[CHEAP_ANTHROPIC_MODEL])
     resources.defer(lambda: client.delete_key(key))
 
     # 1. exhaust the tight window -> litellm returns budget_exceeded

tests/e2e/quota_management/spend_tracking/conftest.py

@@ -19,6 +19,7 @@ from typing import Iterator
 
 import pytest
 
+from e2e_config import CHEAP_ANTHROPIC_MODEL
 from models import LiteLLMParamsBody
 from spend_e2e_client import SpendClient, build_client
 
@@ -32,7 +33,7 @@ def _driver_params(provider_model: str, env_var: str) -> LiteLLMParamsBody:
 
 DRIVER_MODELS: tuple[tuple[str, str, str], ...] = (
     ("gemini-2.5-flash", "gemini/gemini-2.5-flash", "GEMINI_API_KEY"),
-    ("claude-haiku-4-5", "anthropic/claude-haiku-4-5", "ANTHROPIC_API_KEY"),
+    (CHEAP_ANTHROPIC_MODEL, "anthropic/claude-haiku-4-5", "ANTHROPIC_API_KEY"),
     ("openai-text-embedding-3-small", "openai/text-embedding-3-small", "OPENAI_API_KEY"),
 )
 

tests/e2e/quota_management/spend_tracking/test_spend_tracking_e2e.py

@@ -21,6 +21,7 @@ from concurrent.futures import ThreadPoolExecutor
 
 import pytest
 
+from e2e_config import CHEAP_ANTHROPIC_MODEL
 from e2e_http import Result, Success
 from lifecycle import ResourceManager
 from models import ChatResponse, SpendLogs, SpendLogsParams
@@ -397,14 +398,14 @@ def test_each_model_on_a_shared_key_gets_its_own_row(
     )
     claude = unwrap(
         client.chat(
-            scoped_key, "claude-haiku-4-5", f"one word {unique_marker()}", max_tokens=16
+            scoped_key, CHEAP_ANTHROPIC_MODEL, f"one word {unique_marker()}", max_tokens=16
         )
     )
 
     def both_models_costed(rows: list[SpendLogRow]) -> bool:
         costed = [r.model or "" for r in rows if (r.spend or 0) > 0]
         return any("gemini-2.5-flash" in m for m in costed) and any(
-            "claude-haiku-4-5" in m for m in costed
+            CHEAP_ANTHROPIC_MODEL in m for m in costed
         )
 
     rows = client.poll_logs_for_key(scoped_key, min_rows=2, predicate=both_models_costed)
@@ -412,7 +413,7 @@ def test_each_model_on_a_shared_key_gets_its_own_row(
         rows, lambda r: "gemini-2.5-flash" in (r.model or ""), "for the gemini call"
     )
     claude_row = _require_row(
-        rows, lambda r: "claude-haiku-4-5" in (r.model or ""), "for the claude call"
+        rows, lambda r: CHEAP_ANTHROPIC_MODEL in (r.model or ""), "for the claude call"
     )
 
     assert (gemini_row.spend or 0) > 0, f"gemini row should cost > 0: {_summarize(rows)}"