feat(router): add initial support for anthropic messages endpoint

src/vllm_router/app.py

@@ -28,6 +28,7 @@
 )
 from vllm_router.experimental import get_feature_gates, initialize_feature_gates
 from vllm_router.parsers.parser import parse_args
+from vllm_router.routers.anthropic_router import anthropic_router
 from vllm_router.routers.batches_router import batches_router
 from vllm_router.routers.files_router import files_router
 from vllm_router.routers.main_router import main_router
@@ -321,6 +322,7 @@ def initialize_all(app: FastAPI, args):
 
 app = FastAPI(lifespan=lifespan)
 app.include_router(main_router)
+app.include_router(anthropic_router)
 app.include_router(files_router)
 app.include_router(batches_router)
 app.include_router(metrics_router)

src/vllm_router/routers/anthropic_router.py

@@ -0,0 +1,26 @@
+# Copyright 2024-2025 The vLLM Production Stack Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from fastapi import APIRouter, BackgroundTasks, Request
+
+from vllm_router.log import init_logger
+from vllm_router.services.request_service.request import route_general_request
+
+logger = init_logger(__name__)
+anthropic_router = APIRouter()
+
+
+@anthropic_router.post("/v1/messages")
+async def route_anthropic_messages(request: Request, background_tasks: BackgroundTasks):
+    """Route Anthropic-compatible messages requests to the backend."""
+    return await route_general_request(request, "/v1/messages", background_tasks)

src/vllm_router/utils.py

@@ -73,6 +73,7 @@ class ModelType(enum.Enum):
     score = "score"
     transcription = "transcription"
     vision = "vision"
+    messages = "messages"
 
     @staticmethod
     def get_url(model_type: str):
@@ -89,6 +90,8 @@ def get_url(model_type: str):
                 return "/v1/score"
             case ModelType.transcription:
                 return "/v1/audio/transcriptions"
+            case ModelType.messages:
+                return "/v1/messages"
 
     @staticmethod
     def get_test_payload(model_type: str):
@@ -112,6 +115,17 @@ def get_test_payload(model_type: str):
                 return {"query": "Hello", "documents": ["Test"]}
             case ModelType.score:
                 return {"encoding_format": "float", "text_1": "Test", "test_2": "Test2"}
+            case ModelType.messages:
+                return {
+                    "messages": [
+                        {
+                            "role": "user",
+                            "content": "Hello",
+                        }
+                    ],
+                    "temperature": 0.0,
+                    "max_tokens": 3,
+                }
             case ModelType.transcription:
                 if _SILENT_WAV_BYTES is not None:
                     logger.debug("=====Silent WAV Bytes is being used=====")