test(e2e): point four suites at models the providers still serve (#34567)

* test(e2e): point four suites at models the providers still serve

Four llm_translation tests failed against upstream because the model they name
no longer exists. Each replacement was verified against the live stage proxy.

deepseek/deepseek-reasoner is gone; the DeepSeek API now lists only
deepseek-v4-flash and deepseek-v4-pro. Use deepseek/deepseek-v4-pro, which
still returns message.reasoning_content by default and still drops it for both
reasoning_effort="none" and thinking={"type": "disabled"} (litellm maps the
former to the latter, so the provider rejecting a bare "none" does not matter).

amazon.titan-image-generator-v2:0 returns "This model version has reached the
end of its life"; amazon.nova-canvas-v1:0 is the text-to-image model Bedrock
still offers in us-east-1.

Bedrock's Rerank API requires a full model ARN and rejects a bare model id with
"The provided model ARN for reranking is invalid", regardless of model or
region. Pass the ARN for cohere.rerank-v3-5:0, which is available in the
stack's us-east-1.

vertex_ai/gemini-embedding-2 404s as an unknown publisher model on this
project; vertex_ai/text-embedding-005 returns a vector.

* test(e2e): skip the hosted_vllm chat test when its server is unset

test_hosted_vllm_chat_returns_content read os.environ["HOSTED_VLLM_API_BASE"]
directly, so a stack without that env var failed the test with a bare KeyError
instead of reporting an environment gap. The batches suite already skips on the
same variable, and the vertex passthrough tests use pytest.skip for the same
reason, so follow that idiom here.

Drop the HOSTED_VLLM_API_KEY plumbing: the stage vLLM stand-in serves
/v1/chat/completions unauthenticated, and api_key is optional on
LiteLLMParamsBody, so passing it added nothing.

Default the backend to the model that server actually serves,
Qwen/Qwen2.5-0.5B-Instruct-GGUF:Q4_K_M, rather than a Llama id it never had.
Verified against the live stage proxy: a deployment with just that model and
api_base returns "hello".

* fix(model_map): mark deepseek v4-pro and v4-flash as reasoning-capable

Review on #34567 flagged that deepseek/deepseek-v4-pro is not marked
reasoning-capable while the e2e control case requires reasoning_content back
from it. The behavior premise is inverted, but it surfaced a real data gap: the
model map never gained supports_reasoning for the v4 models when DeepSeek
retired deepseek-reasoner, which did carry the flag.

Both models do reason. Against the live API with no reasoning params, v4-pro
returns 106 chars of reasoning_content and v4-flash returns 54, and both drop
it for thinking={"type":"disabled"}.

The stale flag had a real consequence beyond metadata: DeepSeekChatConfig
._thinking_mode_active() gates on supports_reasoning(), so with the flag unset
it returned False even when a caller passed thinking={"type": "enabled"},
skipping the multi-turn check that reasoning_content be passed back on
assistant messages. Param support itself was never gated, which is why
reasoning_effort="none" still mapped to thinking disabled.

Verified with LITELLM_LOCAL_MODEL_COST_MAP=True: supports_reasoning now
reports True for deepseek/deepseek-v4-pro and deepseek/deepseek-v4-flash.

tencent/deepseek-v4-pro is left alone; that route was not exercised here.

litellm/model_prices_and_context_window_backup.json

@@ -45890,6 +45890,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,
@@ -45915,6 +45916,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,
@@ -45940,6 +45942,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,
@@ -45999,6 +46002,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,

model_prices_and_context_window.json

@@ -46012,6 +46012,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,
@@ -46037,6 +46038,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,
@@ -46062,6 +46064,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,
@@ -46087,6 +46090,7 @@
         "supports_native_streaming": true,
         "supports_parallel_function_calling": true,
         "supports_prompt_caching": true,
+        "supports_reasoning": true,
         "supports_response_schema": true,
         "supports_system_messages": true,
         "supports_tool_choice": true,

tests/e2e/llm_translation/test_chat_completions_regression_e2e.py

@@ -343,10 +343,14 @@ class TestHostedVllmChat:
     def test_hosted_vllm_chat_returns_content(
         self, client: PassthroughClient, resources: ResourceManager
     ) -> None:
-        api_base = os.environ["HOSTED_VLLM_API_BASE"]
-        api_key = (os.environ.get("HOSTED_VLLM_API_KEY") or "").strip() or None
+        api_base = os.environ.get("HOSTED_VLLM_API_BASE")
+        if api_base is None:
+            pytest.skip(
+                "set HOSTED_VLLM_API_BASE (the live vLLM server this deployment targets)"
+            )
         backend = (
-            os.environ.get("HOSTED_VLLM_MODEL") or "meta-llama/Llama-3.2-3B-Instruct"
+            os.environ.get("HOSTED_VLLM_MODEL")
+            or "Qwen/Qwen2.5-0.5B-Instruct-GGUF:Q4_K_M"
         ).strip()
         model = f"e2e-vllm-chat-{unique_marker()}"
         model_id = client.proxy.create_model(
@@ -354,7 +358,6 @@ class TestHostedVllmChat:
             LiteLLMParamsBody(
                 model=f"hosted_vllm/{backend}",
                 api_base=api_base,
-                api_key=api_key,
             ),
         )
         resources.defer(lambda: client.proxy.delete_model(model_id))

tests/e2e/llm_translation/test_deepseek_reasoning_e2e.py

@@ -5,7 +5,7 @@ DeepSeek's reasoner defaults thinking ON and surfaces the chain as
 ``reasoning_effort="none"`` and ``thinking={"type": "disabled"}``. The DeepSeek
 param mapper (``litellm/llms/deepseek/chat/transformation.py``
 ``map_openai_params``) forwards both as ``thinking={"type": "disabled"}`` so the
-outbound body carries a real disable signal and ``deepseek-reasoner`` returns no
+outbound body carries a real disable signal and the reasoning model returns no
 ``reasoning_content``. This is the behavior tracked by LIT-3686 / GH #27453.
 
 The control case proves the model and path work (reasoning is returned when
@@ -27,7 +27,7 @@ from passthrough_client import PassthroughClient
 
 pytestmark = pytest.mark.e2e
 
-REASONER = "deepseek/deepseek-reasoner"
+REASONER = "deepseek/deepseek-v4-pro"
 PROMPT = "What is 17 + 26? Answer with just the number."
 
 
@@ -67,7 +67,7 @@ class TestDeepSeekReasoningDisable:
         )
         reasoning = _reasoning_content(response)
         assert reasoning, (
-            "control case: deepseek-reasoner returned no reasoning_content with no "
+            "control case: the reasoning model returned no reasoning_content with no "
             f"disable param, so the disable assertions below can't be trusted: {response}"
         )
 

tests/e2e/llm_translation/test_embeddings_endpoint_e2e.py

@@ -72,7 +72,7 @@ class TestEmbeddingsEndpoint:
         model_id = endpoints_client.create_model(
             model,
             LiteLLMParamsBody(
-                model="vertex_ai/gemini-embedding-2",
+                model="vertex_ai/text-embedding-005",
                 vertex_project="os.environ/VERTEXAI_PROJECT",
                 vertex_location="us-central1",
             ),

tests/e2e/llm_translation/test_image_generation_e2e.py

@@ -54,7 +54,7 @@ class TestImageGeneration:
         model_id = endpoints_client.create_model(
             model,
             LiteLLMParamsBody(
-                model="bedrock/amazon.titan-image-generator-v2:0",
+                model="bedrock/amazon.nova-canvas-v1:0",
                 aws_access_key_id="os.environ/AWS_ACCESS_KEY_ID",
                 aws_secret_access_key="os.environ/AWS_SECRET_ACCESS_KEY",
                 aws_region_name="os.environ/AWS_REGION",

tests/e2e/llm_translation/test_rerank_e2e.py

@@ -60,7 +60,7 @@ class TestRerank:
         model_id = endpoints_client.create_model(
             model,
             LiteLLMParamsBody(
-                model="bedrock/amazon.rerank-v1:0",
+                model="bedrock/arn:aws:bedrock:us-east-1::foundation-model/cohere.rerank-v3-5:0",
                 aws_access_key_id="os.environ/AWS_ACCESS_KEY_ID",
                 aws_secret_access_key="os.environ/AWS_SECRET_ACCESS_KEY",
                 aws_region_name="os.environ/AWS_REGION",