chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

Revert "tiny remove deprecated endpoint call" (#14533)

Browse Source
This commit is contained in:
Baizhou Zhang
2025-12-05 23:48:54 -08:00
committed by GitHub
parent d257bf87b9
commit 42fcf5438f
36 changed files with 59 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/srt/ep/test_deepep_large.py
+1 -1
View File
@@ -143,7 +143,7 @@ class TestDeepseekMTP(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.92)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/ep/test_deepep_small.py
+2 -2
View File
@@ -307,7 +307,7 @@ class TestMTP(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
@@ -381,7 +381,7 @@ class TestMTPWithTBO(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/hicache/test_hicache_variants.py
+1 -1
View File
@@ -152,7 +152,7 @@ class TestHiCacheEagle(HiCacheBaseServer, HiCacheEvalMixin):
self.assertGreaterEqual(metrics["score"], self.expected_mmlu_score)
# EAGLE-specific check
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
print(f"{server_info=}")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
test/srt/quant/test_w4a8_deepseek_v3.py
+1 -1
View File
@@ -103,7 +103,7 @@ class TestDeepseekV3W4Afp8Mtp(CustomTestCase):
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/rl/test_update_weights_from_disk.py
+3 -3
View File
@@ -113,7 +113,7 @@ class TestServerUpdateWeightsFromDisk(CustomTestCase):
return response.json()
def get_model_info(self):
response = requests.get(self.base_url + "/model_info")
response = requests.get(self.base_url + "/get_model_info")
model_path = response.json()["model_path"]
print(json.dumps(response.json()))
return model_path
@@ -254,7 +254,7 @@ class TestServerUpdateWeightsFromDiskAbortAllRequests(CustomTestCase):
return response.json()
def get_model_info(self):
response = requests.get(self.base_url + "/model_info")
response = requests.get(self.base_url + "/get_model_info")
model_path = response.json()["model_path"]
print(json.dumps(response.json()))
return model_path
@@ -386,7 +386,7 @@ class TestUpdateWeightsFromDiskParameterized(CustomTestCase):
return response.json()["text"]
def get_model_info():
response = requests.get(base_url + "/model_info")
response = requests.get(base_url + "/get_model_info")
model_path = response.json()["model_path"]
print(json.dumps(response.json()))
return model_path
test/srt/rl/test_update_weights_from_tensor.py
+1 -1
View File
@@ -209,7 +209,7 @@ class TestServerUpdateWeightsFromTensorNonBlocking(CustomTestCase):
return response.json()
def get_model_info(self):
response = requests.get(self.base_url + "/model_info")
response = requests.get(self.base_url + "/get_model_info")
model_path = response.json()["model_path"]
print(json.dumps(response.json()))
return model_path
test/srt/test_data_parallelism.py
+2 -2
View File
@@ -65,12 +65,12 @@ class TestDataParallelism(CustomTestCase):
def test_get_memory_pool_size(self):
# use `get_server_info` instead since `get_memory_pool_size` is merged into `get_server_info`
response = requests.get(self.base_url + "/server_info")
response = requests.get(self.base_url + "/get_server_info")
assert response.status_code == 200
time.sleep(1)
response = requests.get(self.base_url + "/server_info")
response = requests.get(self.base_url + "/get_server_info")
assert response.status_code == 200
test/srt/test_deepseek_v32_mtp.py
+1 -1
View File
@@ -69,7 +69,7 @@ class TestDeepseekV32MTP(CustomTestCase):
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_deepseek_v3_fp4_4gpu.py
+1 -1
View File
@@ -143,7 +143,7 @@ class TestDeepseekV3FP4MTP(CustomTestCase):
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_deepseek_v3_mtp.py
+1 -1
View File
@@ -67,7 +67,7 @@ class TestDeepseekV3MTP(CustomTestCase):
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_dp_attention.py
+1 -1
View File
@@ -112,7 +112,7 @@ class TestDPAttentionDP2TP2DeepseekV3MTP(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_eagle_dp_attention.py
+1 -1
View File
@@ -78,7 +78,7 @@ class TestEAGLE3EngineDPAttention(CustomTestCase):
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
server_data = server_info.json()
# Try to get avg_spec_accept_length
test/srt/test_eagle_infer_b.py
+1 -1
View File
@@ -144,7 +144,7 @@ class TestEAGLEServer(CustomTestCase):
print(f"{metrics=}")
self.assertGreater(metrics["accuracy"], 0.20)
server_info = requests.get(self.base_url + "/server_info").json()
server_info = requests.get(self.base_url + "/get_server_info").json()
avg_spec_accept_length = server_info["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_fa3.py
+1 -1
View File
@@ -112,7 +112,7 @@ class BaseFlashAttentionTest(CustomTestCase):
self.assertGreater(metrics[metric_key], self.accuracy_threshold)
if self.speculative_decode:
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_flashmla.py
+1 -1
View File
@@ -119,7 +119,7 @@ class TestFlashMLAMTP(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
print(f"{server_info=}")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
test/srt/test_hybrid_attn_backend.py
+1 -1
View File
@@ -88,7 +88,7 @@ class TestHybridAttnBackendBase(CustomTestCase):
self.assertGreater(metrics[metric_key], self.accuracy_threshold)
if self.speculative_decode:
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_mla_deepseek_v3.py
+1 -1
View File
@@ -179,7 +179,7 @@ class TestDeepseekV3MTP(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_mla_flashinfer.py
+1 -1
View File
@@ -111,7 +111,7 @@ class TestFlashinferMLAMTP(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
print(f"{server_info=}")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
test/srt/test_mla_int8_deepseek_v3.py
+2 -2
View File
@@ -113,7 +113,7 @@ class TestDeepseekV3MTPChannelInt8(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
@@ -217,7 +217,7 @@ class TestDeepseekV3MTPBlockInt8(CustomTestCase):
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_ngram_speculative_decoding.py
+1 -1
View File
@@ -80,7 +80,7 @@ class TestNgramSpeculativeDecodingBase(CustomTestCase):
metric_key = "accuracy"
self.assertGreater(metrics[metric_key], self.accuracy_threshold)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
test/srt/test_srt_endpoint.py
+2 -2
View File
@@ -496,7 +496,7 @@ class TestSRTEndpoint(CustomTestCase):
self.assertEqual(send_and_check_cached_tokens(range(0, 11000)), 10000)
def test_get_server_info(self):
response = requests.get(self.base_url + "/server_info")
response = requests.get(self.base_url + "/get_server_info")
response_json = response.json()
max_total_num_tokens = response_json["max_total_num_tokens"]
@@ -626,7 +626,7 @@ class TestSRTEndpoint(CustomTestCase):
tp = ThreadPoolExecutor(max_workers=30)
def s():
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
server_info.json()
futures = []
test/srt/test_standalone_speculative_decoding.py
+1 -1
View File
@@ -88,7 +88,7 @@ class TestStandaloneSpeculativeDecodingBase(CustomTestCase):
metric_key = "accuracy"
self.assertGreater(metrics[metric_key], self.accuracy_threshold)
server_info = requests.get(self.base_url + "/server_info")
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]