Organize public APIs (#809)

python/sglang/__init__.py

@@ -22,46 +22,53 @@ from sglang.api import (
     video,
 )
 
+# SGLang DSL APIs
+__all__ = [
+    "Runtime",
+    "assistant",
+    "assistant_begin",
+    "assistant_end",
+    "flush_cache",
+    "function",
+    "gen",
+    "gen_int",
+    "gen_string",
+    "get_server_args",
+    "image",
+    "select",
+    "set_default_backend",
+    "system",
+    "system_begin",
+    "system_end",
+    "user",
+    "user_begin",
+    "user_end",
+    "video",
+]
+
+
 # Global Configurations
 from sglang.global_config import global_config
 
+__all__ += ["global_config"]
+
 # SGL Backends
 from sglang.lang.backend.runtime_endpoint import RuntimeEndpoint
 from sglang.utils import LazyImport
-from sglang.version import __version__
 
 Anthropic = LazyImport("sglang.lang.backend.anthropic", "Anthropic")
 LiteLLM = LazyImport("sglang.lang.backend.litellm", "LiteLLM")
 OpenAI = LazyImport("sglang.lang.backend.openai", "OpenAI")
 VertexAI = LazyImport("sglang.lang.backend.vertexai", "VertexAI")
 
+__all__ += ["RuntimeEndpoint", "Anthropic", "LiteLLM", "OpenAI", "VertexAI"]
 
-# public APIs management
-__all__ = [
-    "global_config",
-    "Anthropic",
-    "LiteLLM",
-    "OpenAI",
-    "RuntimeEndpoint",
-    "VertexAI",
-    "function",
-    "Runtime",
-    "set_default_backend",
-    "flush_cache",
-    "get_server_args",
-    "gen",
-    "gen_int",
-    "gen_string",
-    "image",
-    "video",
-    "select",
-    "system",
-    "user",
-    "assistant",
-    "user_begin",
-    "user_end",
-    "assistant_begin",
-    "assistant_end",
-    "system_begin",
-    "system_end",
-]
+# Version
+from sglang.version import __version__
+
+__all__ += ["__version__"]
+
+# Core Benchmarks
+from sglang.benchmarks import bench_latency, bench_serving
+
+__all__ += ["bench_latency", "bench_serving"]

python/sglang/benchmarks/__init__.py

@@ -0,0 +1 @@
+"""SGLang core benchmarks."""

python/sglang/benchmarks/bench_latency.py

@@ -2,10 +2,10 @@
 Benchmark the latency of a given model. It accepts arguments similar to those of launch_server.py.
 
 # Usage (latency test):
-python -m sglang.bench_latency --model-path meta-llama/Meta-Llama-3-8B-Instruct --load-format dummy
+python -m sglang.benchmarks.bench_latency --model-path meta-llama/Meta-Llama-3-8B-Instruct --load-format dummy
 
 # Usage (correctness test):
-python -m sglang.bench_latency --model-path TinyLlama/TinyLlama-1.1B-Chat-v0.4 --correct
+python -m sglang.benchmarks.bench_latency --model-path TinyLlama/TinyLlama-1.1B-Chat-v0.4 --correct
 
 ### Reference output:
 prefill logits (first half) tensor([[-10.0312,  -9.5000,   0.8936,  ...,  -4.9414,  -3.2402,  -3.3633],

python/sglang/benchmarks/bench_serving.py

@@ -4,10 +4,10 @@
 Benchmark online serving.
 
 Usage:
-python3 -m sglang.bench_serving --backend sglang --num-prompt 10
+python3 -m sglang.benchmarks.bench_serving --backend sglang --num-prompt 10
 
-python3 -m sglang.bench_serving --backend sglang --dataset-name random --num-prompts 3000 --random-input 1024 --random-output 1024 --random-range-ratio 0.5
-python3 -m sglang.bench_serving --backend sglang --dataset-name random --request-rate-range 1,2,4,8,16,32 --random-input 4096 --random-output 1024 --random-range-ratio 0.125 --multi
+python3 -m sglang.benchmarks.bench_serving --backend sglang --dataset-name random --num-prompts 3000 --random-input 1024 --random-output 1024 --random-range-ratio 0.5
+python3 -m sglang.benchmarks.bench_serving --backend sglang --dataset-name random --request-rate-range 1,2,4,8,16,32 --random-input 4096 --random-output 1024 --random-range-ratio 0.125 --multi
 """
 
 import argparse