Migrate all callers from /get_server_info to /server_info (#21463)

python/sglang/bench_serving.py

@@ -1402,7 +1402,7 @@ async def benchmark(
 
     if "sglang" in backend:
         server_info = requests.get(
-            base_url + "/get_server_info", headers=get_auth_headers()
+            base_url + "/server_info", headers=get_auth_headers()
         )
         if server_info.status_code == 200:
             server_info_json = server_info.json()
@@ -1538,7 +1538,7 @@ async def benchmark(
         print("{:<40} {:<10.2f}".format("Max ITL (ms):", metrics.max_itl_ms))
     print("=" * 50)
 
-    resp = requests.get(base_url + "/get_server_info", headers=get_auth_headers())
+    resp = requests.get(base_url + "/server_info", headers=get_auth_headers())
     server_info = resp.json() if resp.status_code == 200 else None
 
     if (

python/sglang/lang/backend/runtime_endpoint.py

@@ -67,7 +67,7 @@ class RuntimeEndpoint(BaseBackend):
 
     def get_server_info(self):
         res = http_request(
-            self.base_url + "/get_server_info",
+            self.base_url + "/server_info",
             api_key=self.api_key,
             verify=self.verify,
         )
@@ -531,7 +531,7 @@ class Runtime:
 
     async def get_server_info(self):
         async with aiohttp.ClientSession() as session:
-            async with session.get(f"{self.url}/get_server_info") as response:
+            async with session.get(f"{self.url}/server_info") as response:
                 if response.status == 200:
                     return await response.json()
                 else:

python/sglang/profiler.py

@@ -42,7 +42,7 @@ def run_profile(
     # Dump server args.
     file_path = Path(output_dir) / "server_args.json"
     if not file_path.exists():
-        response = requests.get(url + "/get_server_info")
+        response = requests.get(url + "/server_info")
         response.raise_for_status()
         server_args_data = response.json()
         with open(file_path, "w") as file:

python/sglang/test/bench_one_batch_server_internal.py

@@ -609,7 +609,7 @@ def run_one_case(
         last_gen_throughput = -1
         acc_length = -1
     else:
-        response = requests.get(url + "/get_server_info", timeout=DEFAULT_TIMEOUT)
+        response = requests.get(url + "/server_info", timeout=DEFAULT_TIMEOUT)
         response.raise_for_status()
         server_info = response.json()
         internal_state = server_info.get("internal_states", [{}])
@@ -793,7 +793,7 @@ def run_benchmark_internal(
         skip_max_running_requests_threshold = float("inf")
     else:
         model_name = None
-        response = requests.get(base_url + "/get_server_info", timeout=DEFAULT_TIMEOUT)
+        response = requests.get(base_url + "/server_info", timeout=DEFAULT_TIMEOUT)
         response.raise_for_status()
         server_info = response.json()
         if "tokenizer_path" in server_info:

python/sglang/test/kits/cache_hit_kit.py

@@ -221,7 +221,7 @@ async def _send_round(
 def _get_page_size(base_url: str) -> int:
     """Query server for page_size used by radix cache."""
     try:
-        resp = requests.get(f"{base_url}/get_server_info", timeout=10)
+        resp = requests.get(f"{base_url}/server_info", timeout=10)
         resp.raise_for_status()
         info = resp.json()
         return info.get("page_size", 1)

python/sglang/test/kl_test_utils.py

@@ -208,7 +208,7 @@ def test_input_output_logprobs_match_helper(
 def test_input_output_logprobs_match_prefill_cache_hit_helper(
     base_url, ACC_THRESHOLDS, model_name, max_samples=None, max_new_tokens=8192
 ):
-    server_info = requests.get(base_url + "/get_server_info").json()
+    server_info = requests.get(base_url + "/server_info").json()
     if server_info["disable_radix_cache"]:
         print("Radix cache is disabled, skipping test")
         return
@@ -261,7 +261,7 @@ def test_input_output_logprobs_match_prefill_cache_hit_helper(
 def test_input_output_logprobs_match_decode_cache_hit_helper(
     base_url, ACC_THRESHOLDS, model_name, max_samples=None, max_new_tokens=8192
 ):
-    server_info = requests.get(base_url + "/get_server_info").json()
+    server_info = requests.get(base_url + "/server_info").json()
     if server_info["disable_radix_cache"]:
         print("Radix cache is disabled, skipping test")
         return

python/sglang/test/nightly_utils.py

@@ -324,7 +324,7 @@ class NightlyBenchmarkRunner:
             The average speculative decoding accept length, or None if not available.
         """
         try:
-            response = requests.get(f"{self.base_url}/get_server_info", timeout=10)
+            response = requests.get(f"{self.base_url}/server_info", timeout=10)
             if response.status_code == 200:
                 server_info = response.json()
                 internal_states = server_info.get("internal_states", [])