diff --git a/tests/e2e/llm_translation/test_chat_completions_regression_e2e.py b/tests/e2e/llm_translation/test_chat_completions_regression_e2e.py
index 6a69384d31a..655d426c28d 100644
--- a/tests/e2e/llm_translation/test_chat_completions_regression_e2e.py
+++ b/tests/e2e/llm_translation/test_chat_completions_regression_e2e.py
@@ -84,12 +84,6 @@ OPENAI_VISION_BACKEND = "openai/gpt-4o"
 
 # OpenAI caches a shared prompt prefix once it exceeds ~1024 tokens; this is well
 # past that, so a repeat call reports cached prompt tokens.
-CACHE_PREFIX = (
-    "You are a meticulous assistant. Follow these standing instructions exactly. "
-    * 300
-)
-
-
 def _vision_messages() -> list[ChatMessage]:
     return [
         ChatMessage(
@@ -582,36 +576,6 @@ class TestOpenAIChatCompletions:
         response = unwrap(client.proxy.chat(key, ChatBody(model=model, messages=_vision_messages(), max_tokens=32)))
         _assert_describes_cat(response)
 
-    @pytest.mark.covers(
-        "llm.chat_completions.openai.prompt_cache_5m.nonstream.works",
-        exercised_on=["chat_completions"],
-    )
-    def test_openai_chat_prompt_cache_hits_on_repeat(
-        self, client: PassthroughClient, resources: ResourceManager
-    ) -> None:
-        model = f"e2e-openai-cache-{unique_marker()}"
-        model_id = client.proxy.create_model(
-            model, LiteLLMParamsBody(model=OPENAI_BACKEND, api_key="os.environ/OPENAI_API_KEY")
-        )
-        resources.defer(lambda: client.proxy.delete_model(model_id))
-        key = resources.key()
-
-        body = ChatBody(
-            model=model,
-            messages=[
-                ChatMessage(role="system", content=CACHE_PREFIX),
-                ChatMessage(role="user", content="Reply with the single word pong."),
-            ],
-            max_tokens=16,
-        )
-        unwrap(client.proxy.chat(key, body))
-        second = unwrap(client.proxy.chat(key, body))
-
-        details = second.usage.prompt_tokens_details if second.usage else None
-        assert details and details.cached_tokens and details.cached_tokens > 0, (
-            f"a repeated large-prefix prompt must report cached prompt tokens, got usage={second.usage}"
-        )
-
     @pytest.mark.covers(
         "llm.chat_completions.openai.tool_use.stream.works",
         exercised_on=["chat_completions"],